the effects of planning conditions on narrative …
TRANSCRIPT
THE EFFECTS OF PLANNING CONDITIONS ON
CHINESE PRIMARY SCHOOL ESL PUPILS’
NARRATIVE WRITING
MA FEI FAN
DISSERTATION SUBMITTED IN PARTIAL
FULFILMENT OF THE REQUIREMENTS FOR THE
DEGREE OF MASTER OF
ENGLISH AS A SECOND LANGUAGE
FACULTY OF LANGUAGE AND LINGUISTICS
UNIVERSITY OF MALAYA
KUALA LUMPUR
2016
Univers
ity of
Mala
ya
ii
UNIVERSITY OF MALAYA
ORIGINAL LITERARY WORK DECLARATION
Name of Candidate: MA FEI FAN I.C. No.:
Registration/ Matric No.: TGB120001
Name of Degree: Master of English as a Second Language
Title of Dissertation (“this Work”):
The Effects of Planning Conditions on Chinese Primary School ESL Pupils’
Narrative Writing
Field of Study: Second Language Acquisition
I do solemnly and sincerely declare that:
(1) I am the sole author/ writer of this work;
(2) This work is original;
(3) Any use of my work in which copyright exists was done by way of fair dealing and for
permitted purposes and any expert or exact from, or reference to or reproduction of any
copyright work has been disclosed expressly and sufficiently and the title of the Work
and its authorship have been acknowledged by this Work;
(4) I do not have any actual knowledge nor do I ought reasonably to know that the making
of this work constitutes an infringement of any copyright work;
(5) I hereby assign all and every rights in the copyright to this Work to the University of
Malaya (“UM”), who henceforth shall be owner of the copyright in this Work and that
reproduction or use in any form or by any means whatsoever is prohibited without the
written consent of UM having been first had and obtained;
(6) I am fully aware that if in the course of making this Work I have infringed any
copyright whether intentionally or otherwise, I may be object to legal action or any
other action as may be determined by UM.
Candidate’s Signature: Date:
Subscribed and solemnly declared before,
Witness’s Signature Date:
Name :
Designation :
Univers
ity of
Mala
ya
iii
ABSTRACT
The present study investigates how 78 pupils from a Chinese primary school in the state
of Selangor performed differently in their second language writing under different
planning conditions in terms of the language complexity, accuracy and fluency (CAF).
Besides, the study also aims to examine the pupils’ perceptions of their own
performance under different planning conditions and how these perceptions are related
to their performance. Drawing from Kellogg’s (1996) model as the theoretical
framework, this study also seeks to describe how pupils perform their narrative writing
task mentally. The pupils were divided into three planning condition groups:1) pre-task
planning, 2) on-line planning and 3) no planning groups. A variety of data collection
methods had been employed in order to obtain pupils’ data. These methods include the
collection of pupils’ written task and pupils’ planning sheets, questionnaire surveys and
interview. The results of the study reveal that pupils’ performance was not significantly
affected by the planning conditions except for the effect of pre-task planning on pupils’
written fluency. The results discovered from the present study differ greatly from many
of the previous studies which were mainly carried out on adult writers. The present
study also discovers that pupils’ perceptions partially influence their performance. The
reasons of such contradictions are discussed in the discussion section in this article.
Keywords: second language writing, complexity, accuracy, fluency, Kellogg’s (1996)
model. Univers
ity of
Mala
ya
iv
ABSTRAK
Kajian ini mengkaji bagaimana 78 orang murid dari sebuah sekolah jenis kebangsaan
Cina di Negeri Selangor menghasilkan penulisan Bahasa Inggeris sebagai bahasa kedua
yang berbeza prestasi di dalam keadaan perancangan penulisan yang berbeza. Prestasi
mereka dinilai berdasarkan kerumitan, ketepatan dan kefasihan bahasa (CAF). Selain itu,
kajian ini menyelidik persepsi murid-murid terhadap prestasi mereka dalam keadaan
perancangan yang berbeza serta kesannya terhadap prestasi mereka dalam penulisan.
Dengan menggunakan model Kellogg (1996) sebagai kerangka teoretikal, kajian ini
juga dijalankan untuk menjelaskan operasi mental yang ada pada murid-murid semasa
menulis. Dalam kajian ini, murid-murid telah dibahagikan kepada tiga kumpulan: 1)
kumpulan perancangan pra-tugasan, 2) kumpulan perancangan semasa tugasan dan 3)
kumpulan kawalan. Kajian ini juga menggunakan pelbagai kaedah pengumpulan data,
di antaranya seperti pengumpulan penulisan murid dan kertas perancangan murid, soal
selidik dan temu bual. Keputusan kajian ini menunjukkan bahawa prestasi murid tidak
terjejas dengan keadaan perancangan yang dialami, kecuali bagi kesan perancangan pra-
tugasan terhadap kefasihan penulisan murid. Keputusan yang diperoleh daripada kajian
ini juga berbeza dengan keputusan yang diperoleh daripada kajian masa lampau, di
mana kebanyakan kajian tersebut dijalankan dalam kalangan orang dewasa. Kajian ini
juga mendapati bahawa persepsi murid juga mempengaruhi sebahagian daripada
prestasi murid. Sebab-sebab yang menyebabkan percanggahan dapatan kajian ini
dengan kajian lampau dibincang dalam bab perbincangan dalam desertasi ini.
Kata Kunci: penulisan bahasa kedua, kerumitan, ketepatan, kefasihan, model Kellogg
(1996)
Univers
ity of
Mala
ya
v
ACKNOWLEDGEMENTS
For the completion of this dissertation, I owe a great debt of gratitude to my supervisor,
Dr. Azlin Zaiti Binti Zainal, for giving me her relentless support and professional
guidance. I am glad that I have learned a lot in the process of completing this
dissertation under her supervision.
Other than that, I would like to express my gratitude to Madam Ainun Rozana
and Dr. Siti Zaidah Binti Zainuddin for their useful feedback. Their recommendations
gave me clear direction on improving my present study.
Lastly, I would like to thank my family members and friends who had given me
mental support especially at the lows of my preparation process. Without them, the
process of completing this dissertation would not have been done so smoothly.
Univers
ity of
Mala
ya
vi
TABLE OF CONTENTS
ORIGINAL LITERARY WORK DECLARATION………………………. ii
ABSTRACT…………………………………………………………………… iii
ABSTRAK…………………………………………………………………….. iv
ACKNOWLEDGEMENTS………………………………………………….. v
TABLE OF CONTENTS…………………………………………………….. vi
LIST OF FIGURES…………………………………………………………... xi
LIST OF TABLES……………………………………………………………. xii
LIST OF ABBREVIATIONS………………………………………………... xiv
CHAPTER ONE: INTRODUCTION……………………………………….. 1
1.0 Introduction…………….……………………………………..................... 1
1.1 Background of the Study………………………………………………… 1
1.1.1 Narrative writing in Chinese primary school curriculum…. 2
1.1.2 The importance of learning narrative writing……………… 4
1.1.3 Task-based language learning……………………………...... 5
1.1.4 Challenges faced by young ESL writers…………………...... 6
1.1.5 Importance of planning to L2 writing………………….…… 6
1.2 Statement of the problem………………………………………………... 7
1.3 Purposes of the study...……………………………………….................... 10
1.4 Objectives of the study…………………………………………………... 10
1.5 Research questions…………………………………………...................... 11
1.6 Significance of the study………………………………………………….. 11
1.7 Scope and limitations of the study……………………………………….. 13
1.8 Organisation of the dissertation…………………………………………. 14
1.9 Conclusion………………………………………………………………… 15
Univers
ity of
Mala
ya
vii
CHAPTER TWO: LITERATURE REVIEW……………………………… 16
2.0 Introduction……………………………………………………………….. 16
2.1 Narrative writing ………………………………………………………… 16
2.2 Second language writing………………………………………………… 17
2.2.1 Measurement of L2 performance: Complexity, accuracy
and fluency (CAF)……………………………………………………..
18
2.3 Planning in second language writing…………………………………… 21
2.3.1 Planning in L2 writing: Theoretical framework……………. 23
2.3.2 Working memory in writing…………………………………. 32
2.3.3 Roles of visuo-spatial sketchpad (VSSP) in writing………... 33
2.3.4 Roles of central executive unit in writing…………………… 34
2.3.5 Roles of the phonological loop in writing………………….... 35
2.3.6 Assumptions in human information processing and their
links to Kellogg’s (1996) model……………………………………….
36
2.3.7 Reasons of selecting Kellogg’s (1996) model as theoretical
framework………………………………………………………….....
37
2.4 Relevant studies on L2 language production performance……………. 39
2.5 Research related to learners’ perception and task performance……… 49
2.6 Identified research gaps………………………………………………… 50
2.7 Conclusion………………………………………………………………… 52
CHAPTER THREE: METHODOLOGY…………………………………... 53
3.0 Introduction……………………………………………………………….. 53
3.1 The pilot study……………………………………………………………. 53
3.1.1 Participants in the pilot study……………………………… 54
3.1.2 Procedures of the pilot study………………………………… 54
3.1.3 Results of the pilot study……………………………………... 56
Univers
ity of
Mala
ya
viii
3.1.4 Suggestions for the main study………………………………. 57
3.2 The main study……………………………………………………………. 57
3.2.1 The participants………………………………………………. 57
3.2.2 Variables……………………………………………………… 59
3.2.3 Research design……………………………………………….. 62
3.2.4 The task………………………………………………………... 65
3.2.5 Instruments used in the study/ data collection methods…… 66
3.2.6 Data analysis methods………………………………………... 67
3.2.7 Addressing the Validity and Reliability Issues of the
Present Study………………………………………………………….
69
3.3 Conclusion………………………………………………………………… 70
CHAPTER FOUR: RESULTS OF PLANNING EFFECTS………………. 71
4.0 Introduction……………………………………………………………….. 71
4.1 Results of independent variables………………………………………… 71
4.2 Results of dependent variables…………………………………………... 73
4.2.1 Complexity…………………………………………………….. 73
4.2.2 Accuracy………………………………………………………. 78
4.2.3 Fluency……………………………………………………… 79
4.3 Results of questionnaire surveys………………………………………… 81
4.4 Results of interview……………………………………………………….. 87
4.5 Analysis of pupils’ planning sheets……………………………………… 97
4.6 Conclusion ………………………………………………………………... 99
CHAPTER FIVE: DISCUSSION ON PLANNING EFFECTS…………… 100
5.0 Introduction……………………………………………………………….. 100
5.1 The effects of planning on Chinese primary school pupils’ written
performance………………………………………………………………….
100
Univers
ity of
Mala
ya
ix
5.1.1 Planning effects on complexity………………………………. 100
5.1.2 Planning effects on accuracy…………………………………. 103
5.1.3 Planning effects on fluency…………………………………... 105
5.1.4 Factors affecting the results of the present study…………... 108
5.2 The ESL writing process of Chinese primary school pupils…………… 113
5.2.1 Trade-off effects……………………………………………… 113
5.2.2 Pupils’ foci in writing under different planning conditions.. 114
5.2.3 Pre-task planning strategies employed by pre-task planners 116
5.3 Chinese primary school pupils’ perceptions on their different
planning conditions……………………………………………………………
118
5.3.1 The differences between pupils’ perceptions and actual
performance……………………………………………………….......
118
5.4 Conclusion………………………………………………………………… 121
CHAPTER SIX: CONCLUSION……………………………………………. 123
6.0 Introduction……………………………………………………………….. 123
6.1 Summary of the present study ………………………………………….. 123
6.2 Theoretical implication of the present study……………………………. 124
6.3 Pedagogical implication of the present study………………………….... 125
6.4 Suggestions for future studies……………………………………………. 126
6.5 Conclusion………………………………………………………………… 129
REFERENCES……………………………………………………………… 130
APPENDIX 1: PILOT STUDY TASK SHEET…………………………….. 148
APPENDIX 2: NOTE EXPANSION TASK FOR THE PTP GROUP……. 150
APPENDIX 3: NOTE EXPANSION TASK FOR THE OLP GROUP…… 152
APPENDIX 4: NOTE EXPANSION TASK FOR THE NP GROUP……... 154
APPENDIX 5: QUESTIONNAIRE FORM 1………………………………. 156
Univers
ity of
Mala
ya
x
APPENDIX 6: QUESTIONNAIRE FORM 2……………………………….
158
APPENDIX 7: INTERVIEW QUESTIONS………………………………...
159
APPENDIX 8: LIST OF CODES USED IN CODING INTERVIEW
TRANSCRIPTS………………………………………………………………
160
Univers
ity of
Mala
ya
xi
LIST OF FIGURES
Figures Title Page
Figure 2.1 Model of Narrative Superstructure (Van Dijk, 1977a) 17
Figure 2.2 Alternative Model of Narrative Superstructure (Kintsch,
1977, as in Cortazzi, 2014)
17
Figure 2.3 Kellogg’s (1996) Model 27
Figure 2.4 Flower and Hayes’ (1981) Model
28
Figure 2.5 Bereiter & Scardamalia’s (1987) Knowledge Transforming
Model of writing
29
Figure 2.6 Model of Writing as Communicative Language Use (Grabe
and Kaplan, 1996)
31
Figure 2.7 Zimmerman’s (2000) Self-regulated Learning Model 32
Figure 3.1 Conceptual Framework for the Design of the Present Study 63
Univers
ity of
Mala
ya
xii
LIST OF TABLES
Tables Title Page
Table 1.1 Topics in UPSR English language papers from 2007 to
2015 (Examination Syndicate, 2013 &2015)
3
Table 3.1 Time allocated for each planning condition 62
Table 3.2 Operationalisation of the writing subprocesses for all
planning conditions
64
Table 4.1 Analysis of independent variables 72
Table 4.2 Analysis of learners’ syntactic complexity 74
Table 4.3 Analysis of learners’ syntactic variety 75
Table 4.4 Analysis of learners’ lexical variety 76
Table 4.5 Percentages of error-free clauses produced by three groups
of pupils
78
Table 4.6 Percentages of correct verb forms produced by three
groups of pupils
79
Table 4.7 Comparison of mean scores of the measure of syllables per
minute
80
Table 4.8 Comparison of mean scores of the measure of word
reformulation
81
Table 4.9 The result of the first part of questionnaire survey: Pupils’
focus in writing
82
Table 4.10 Reliability Statistics for the first part of the questionnaire
survey
83
Table 4.11 Reliability Statistics for the second part of the
questionnaire survey
84
Table 4.12 Pupils’ perception on their performance in terms of
complexity
84
Table 4.13 Pupils’ perception on their performance in terms of
accuracy
86
Table 4.14 Pupils’ perception on their performance in terms of fluency 87
Table 4.15 Mental activities involved by the pupils while performing
the writing task
91
Univers
ity of
Mala
ya
xiii
Table 4.16 Pupils’ feedbacks on their perception of their overall
performance obtained from the interview
91
Table 4.17 Pupils’ perceptions on the CAF of their writing 93
Table 4.18 Types of pre-task planning done by pupils in the PTP
group
97
Table 5.1 Findings and interpretation related to Research Question 1 107
Table 5.2 Findings and interpretation related to Research Question 2
117
Table 5.3 Findings and interpretation related to Research Question 3
121
Univers
ity of
Mala
ya
xiv
LIST OF ABBREVIATIONS
Abbreviation Explanation
ANOVA Analysis of variance
CAF Complexity, accuracy and fluency
ESL English as a second language
KBSR Malaysia’s New Primary School Curriculum
KSSR Primary School Standard Curriculum
L1 First language
L2 Second language
MSTTR Mean segmental type-token ratio
NP No planning
OLP On-line planning
PTP Pre-task planning
RQ Research question
SLA Second language acquisition
SPSS Statistical Package for Social Science
UPSR Primary School Assessment Test
VSSP
Visuospatial sketchpad
Univers
ity of
Mala
ya
1
CHAPTER ONE: INTRODUCTION
1.0 Introduction
The first chapter of the dissertation introduces some important issues regarding the
teaching and practices of writing in the Chinese primary school context. The research
objectives, research questions, problem statements and scope of study are presented in
this chapter.
1.1 Background of the Study
Focusing on the struggle of English language learners in Malaysia in writing, this study
investigates the effects of planning conditions on second language (L2) young writers’
quality of narrative writing. English language, as a lingua franca serves as an important
medium of communication regardless in academia or in daily lives. Malaysia as a
multilingual country is not exempted from the influence of English language. In both
primary and secondary levels, English is taught as a second language and a compulsory
subject in all Malaysian schools (Hiew, 2012). From the British Colonial era until now,
the English language education in Malaysia has experienced a few stages of
transformation. Nevertheless, people have different opinions on the issues of the
changes in language education and often, it is claimed that those transformations
contribute to the falling standards of English language among Malaysians nowadays. In
this chapter, issues of primary English language curricula and their relevance to learners’
writing practices and proficiency will be presented. Understanding these issues is very
important because it provides us with some basic ideas about how writing is taught, how
young learners write, how well they perform in writing, what obstacles they face in
Univers
ity of
Mala
ya
2
writing and also what is the expected written proficiency in Malaysian primary school
contexts.
The present study started in year 2013. It was a year when the primary schools
in Malaysia were undergoing the process of curriculum transformation. The pupils who
studied in Years 4, 5 and 6 learnt English under the KBSR (Malaysia’s New Primary
School Curriculum) syllabus whereas the pupils in Years 1, 2 and 3 studied under the
KSSR (Primary School Standard Curriculum) syllabus. At that time, The KBSR
syllabus was slowly phasing out and would be fully replaced by the KSSR syllabus in
2016 (Ministry of Education Malaysia, 2013). The issues related to the narrative writing
in Chinese primary school under these two curricula will be presented in the next
section.
1.1.1 Narrative Writing in Chinese Primary School Curriculum
In the KBSR syllabus, pupils should be able to acquire a certain number of writing
skills specified in the curriculum specification. For instance, pupils should be able to
write longer texts in the form of paragraphs in guided and/ or free writing, write simple
descriptions of things, events, scene and what one did and saw, and write to express
their feelings and exercise their creativity such as when writing a diary, composing
stories, etc (Ministry of Education Malaysia, 2001). From the aforementioned skills, it
is apparent that narrative writing is an important genre of writing to be taught in the
KBSR English language syllabus.
In the KSSR syllabus, pupils learn English language under three broad themes: 1)
World of Stories, 2) World of Self, Family and Friends and 3) World of Knowledge.
Besides, English language lessons are taught based on five modules which focus on
Univers
ity of
Mala
ya
3
skills such as listening speaking, reading, writing, language arts and grammar (Ministry
of Education Malaysia, 2013). Topics related to narrative writing are usually covered in
writing module under the themes of World of Stories and World of Self, Family and
Friends.
In the UPSR (Primary School Assessment Test) English Language Examination
of the KBSR syllabus, Section C (note expansion) in Paper Two (writing) plays an
integral part in the examination. It contributes 19 percent of the total scores and
narrative writing is the most common genre to be tested in this section. From the
analysis of the actual exam questions in Section C from 2007 to 2015, the researcher
discovered that all questions were of narrating. The analysis was done based on two
reference books which compile actual exam questions in the UPSR English Language
Examination, with the approval from the examination syndicate (Kertas soalan
peperiksaan sebenar 2007-2013 UPSR Bahasa Inggeris; 2013 & Rajathurai, 2016).
Table 1.1 below shows the analysis of the topics in the Section C of the UPSR English
language papers from 2007 to 2015.
Table 1.1. Topics in UPSR English language papers from 2007 to 2015 (Kertas soalan
peperiksaan sebenar 2007-2013 UPSR Bahasa Inggeris; 2013 & Rajathurai, 2016).
Topics 2007 2008 2009 2010 2011 2012 2013 2014 2015
Holidays
Events and Activities / / / / /
Neighbours /
Social Issues /
Incidents / /
In year 2016, all the Year 6 pupils will be taking their examinations using the
new examination formats, as the syllabus of the English language is fully replaced by
the KSSR syllabus. English language will be assessed in two papers: 1) Comprehension
Univers
ity of
Mala
ya
4
and 2) Composition. Moreover, note expansion task will still remain in English
language evaluation and more marks (50 marks) will be allocated to this task (Ministry
of Education Malaysia, 2015). The new examination format in English language
evaluation calls for more attention given to different genres of writing including
narrative writing.
In the practice of teaching writing, pupils in Chinese primary schools are
generally given a lot of time during the week for writing exercises. It is different from
the national schools or Tamil primary schools as these schools focus more on rote
memorisation and repetition in improving writing skills (Lim, 2013). However, in the
UPSR examination, pupils in the Chinese and Tamil primary schools need to sit for
eight subjects whereas pupils in the national schools need to take six subjects. Due to
the number of subjects that they need to cope, it is important for the pupils in primary
schools to master proper English writing skills so that they can pass these assessments.
In the present study, the researcher chose pupils from the Chinese primary schools to be
the participants as the researcher is a Chinese primary school teacher and would like to
use the findings of the present study improve his practice in the instruction of writing.
1.1.2 The Importance of Learning Narrative Writing
There are several reasons to explain why children need to learn narrative writing.
According to Elliot (1995), young writers capture their everyday lives on pages through
narrative writing. The benefits of expressing feeling and experience through writing can
be extended to the psychological and physiological aspects (Graham and Perin, 2007).
Besides, children can be empowered to become critical thinkers by expressing
themselves in narrative writing because children have to use their imagination to write
Univers
ity of
Mala
ya
5
narratives (Burke, 2012). Furthermore, Elliot (1995) pointed out some importance of
narrative writing explicitly, as shown in the following:
a) Narrative writing provides legitimacy: Through writing narratives, children can
find the sense of authenticity in their daily lives.
b) Narrative writing provides metacognition: In order to produce a piece of
narrative, writers have to select and edit events and at the same time “think
about the process of thinking”.
c) Narrative writing provides access to the holy, mysterious and spiritual quality of
human consciousness: Writing narratives helps the awakening of spirit, the
feeling of awe, the idea of mystery and the sensation of fascination.
1.1.3 Task-based Language Learning
In language classrooms, a task is an activity that is conducted using language (Richards
& Rodgers, 2001). According to Patil (2014), tasks should fulfil certain principles of
language learning, such as:
1. Provide the processing of input and output required for language acquisition.
2. Motivate and encourage learners to use authentic language.
3. Task difficulty can be negotiated to suit learning purposes.
In the present study, the task used to obtain pupils’ language output is a note
expansion task. The note expansion task is provided with pictures and keywords for
primary school pupils to write a narrative. The note expansion task used in the present
study fulfils the principles listed by Patil (2014) as it provides keywords for learners to
internalise and at the same time prompts the pupils to write (externalisation). Besides,
the note expansion task encourages pupils to use authentic language as the task is about
Univers
ity of
Mala
ya
6
an event related to their daily lives. Furthermore, with the picture prompts and
keywords given, the difficulty level of the task is already adjusted to the pupils’ level
and at the same time, suits the pupils’ learning needs and appropriate for the purposes of
the present study.
1.1.4 Challenges Faced by Young ESL Writers
For ESL learners, performing a writing task can be challenging because it requires them
to attend to different mental processes. The challenge of completing writing tasks is
even greater for young ESL learners, as they are still developing their proficiency in
English. Young ESL learners have insufficient vocabulary and incomplete mastery of
grammar to complete writing tasks (Nam, 2011). The factors that hinder young ESL
writers’ proficiency include the confusion between school and home environments,
cultural background and also underdeveloped bilingualism (Nam, 2011). Hence, their
writing may not show the complexity of mature writers. They might use whatever
linguistic resource available to them to write and their writing might manifest features
such as invented and unconventional spelling and letter forms, inappropriate
segmentation and punctuation and the use of drawing or writing to express ideas
(Hudelson, 1989).
1.1.5 Importance of Planning to L2 Writing
Planning is one of the sub-processes in writing where the writer establishes goals,
generates ideas and organises those ideas (Ellis, 2005). Due to the attention it attracted
in the studies of second language acquisition (SLA), it is now a field of inquiry on its
own right (Ortega, 2005). From the second language acquisition research to date, there
are two major types of planning conditions: 1) pre-task planning and 2) on-line planning.
Univers
ity of
Mala
ya
7
Pre-task planning is always operationalised as the provision of time given before the
actual performance of a language task (Johnson, Mercado & Acevedo, 2012) whereas
on-line planning is defined as within-task planning where learners are given
unpressured time limit to perform pre-production and post-production monitoring (Ellis
& Yuan, 2004). Planning is important in L2 writing as it may affect how a learner uses
language. According to Skehan and Foster (1999), engagement of different planning
conditions may affect the language production of writers. It is because when writers
plan differently, the cognitive load of the writers will be reduced as the writers engage
in different components of their working memory under a specific planning condition.
Moreover, the manipulation of planning conditions also plays a facilitative role
in interlanguage development (Lin, 2013). From an information processing perspective,
Crookes (1989) postulated that learners push their interlanguage to its limits when they
plan. As the learners use the language which is on the edge of their current linguistic
ability, they engage themselves in second language learning (Crookes, 1989). Thus, it is
possible for language teachers to apply proper manipulation of planning conditions in
L2 writing tasks in order to promote learners’ L2 writing skills.
The beneficial effects of the manipulation of planning conditions on L2 writing
have been found in previous studies. The details of how planning conditions affect
linguistic output, as discovered in those studies will be presented in Chapter 2.
1.2 Statement of the Problem
Writing is one of the pivotal skills to acquire in schooling years but the acquisition of
the skill is challenging. The challenge is even tougher when it comes to the acquisition
of second language writing skills. To relate it to the Malaysian context, the limited
Univers
ity of
Mala
ya
8
proficiency of Malaysian students is one of the issues needs to be solved (Ali, Hamid &
Moni, 2011). In terms of writing skills, having proficient command of writing skills is
crucial in deciding one’s academic success (Graham and Perin, 2007). Thus, in order to
pave the road of success for the students in academic, teaching of the necessary writing
skills should not be taken lightly. In this study, the research examines the effects of
planning, which is a cognitive process in writing, on pupils’ L2 production. As
predicted by Johnson (2012), planning would have positive impact on the complexity,
accuracy and fluency of L2 writing. Furthermore, writers in L2 would be able to focus
more on the translating process (process of selecting grammatical forms and lexical
items) in their writing if they perform planning properly (Oh, Lee & Moon, 2015).
From the observation of the researcher, it is found that majority of pupils in
Chinese primary schools do not achieve the desired competency in performing the note
expansion task in their English language examination paper where in the task, the pupils
should be able to demonstrate their ability to write short descriptions of things, events,
scenes and what they did and saw (Ministry of Education, 2001). This section of the
English language paper is always the part which pupils find most difficult. Other than
that, the researcher also found that pupils do not perform any planning before writing
because the pupils are given keywords in the note expansion task. Moreover, the
keywords given in the note expansion task are arranged according to the order of the
story and at the same time, inflected according to the desired verb forms and therefore,
the provision of keywords hinders the pupils to perform strategic planning prior to the
actual writing. Other than this, based on the researcher’s experience, teachers in Chinese
primary schools do not coach the pupils in performing pre-task planning but they put
more focus on making grammatical sentences and self-checking their own writing. In
the researcher’s opinion, pupils might perform pre-task planning if they are given
proper instruction on how to perform it prior to the examination and it might positively
Univers
ity of
Mala
ya
9
affect pupils’ overall performance. The researcher’s observation is in line with the
statement given by Silva (1993), who reported that L2 writers did very little planning
such as goal setting but spent more effort in generating materials. However, from the
study conducted by Rahimpour and Nariman-Jahan (2011), it was found that planning
could benefit both high and low proficiency learners, as the study claimed that low-
proficiency learners could be benefited in terms of fluency, lexical complexity and
syntactical complexity whilst high proficiency learners could enhance their language
accuracy and lexical complexity (Rahimpour & Nariman-Jahan, 2011). Based on the
advantages mentioned in Rahimpour and Nariman-Jahan’s (2011) study, the present
study is carried out to examine whether the same effects could be yielded from the
Chinese primary school pupils.
Moreover, it is interesting to notice that in the note expansion section of the
English language paper, pupils are required to write a short narrative between 80 to 100
words in average but the time given to perform the task is much more than adequate (35
minutes). The researcher believes that the extra provision of time helps the pupils to
check and improve their writing and reduces the time pressure of the pupils in
performing the writing task. When pupils have extra provision of time, they may be
considered performing the task under on-line planning condition. However, it is still a
doubt whether the extra provision of time is necessary to improve pupils’ writing
quality. If the extra provision of time (on-line planning) does not have any significant
effect in improving or maximising pupils’ quality of writing, the examination syndicate
may need to reconsider the time needed for note expansion tasks, so that the pupils can
use their time more efficiently in attending other tasks in the exam.
Univers
ity of
Mala
ya
10
Hence, there are two identified problems which need to be investigated: 1) the
effects of strategic planning in improving pupils’ writing quality and 2) the effects of
unpressured time condition on pupils’ writing quality.
1.3 Purposes of the Study
Based on the identified problem in second language writing, the study is carried out to
serve several purposes. First, the present study investigates the effects of different
planning conditions (pre-task planning and on-line planning) on the quality of Chinese
primary school pupils in Malaysia. In this research, the planning conditions are adapted
based on Kellogg’s (1996) model. The quality of pupils’ writing is analysed in terms of
complexity, fluency and accuracy (henceforth known as CAF). In addition, the research
hopes to examine how Chinese primary school pupils approach the given note
expansion task. The reason for examining how pupils approach the writing task is to
identify other aspects which may affect pupils’ quality of writing in the writing process.
Moreover, the research also intends to investigate Chinese primary school pupils’
perception about the effects of planning conditions in their writing. The rationale for
investigating pupils’ perception is to identify whether different planning conditions
affect pupils’ psychological state while performing a writing task.
1.4 Objectives of the Study
The research objectives of this study are threefold. First of all, the present study
investigates how planning strategies affect the performance of narrative writing
performed by Chinese primary school pupils. The CAF of the pupils’ written narratives
will be examined. Second, the study investigates how pupils approach the narrative
writing task under different planning condition and to relate the findings to Kellogg’s
Univers
ity of
Mala
ya
11
(1996) Model of Writing Processes and pupils’ performance based on the analysis of
CAF. Third, the study intends to obtain pupils’ opinions of planning strategies in
writing. Pupils’ opinions will be obtained in order to examine how planning affects
pupils’ perception of their performance in terms of CAF.
1.5 Research Questions
Based on the aforementioned objectives, the present study aims to explain how planning
conditions affect Chinese primary school pupils’ narrative writing quality analysed in
terms of CAF. Besides, this study also attempts to find a link between how pupils
approach the narrative writing task under different planning conditions and the actual
performance of the pupils. The study also investigates how the pupils think of their
performance in terms of CAF. Thus, the questions that the present study aims to solve
are:
1. To what extent does planning have on Chinese primary school pupils’ written
complexity, accuracy and fluency when they perform pre-task or on-line planning?
2. How do Chinese primary school pupils approach the narrative writing task under
different planning conditions?
3. What are the Chinese primary school pupils’ perceptions on their performance in
different planning conditions?
1.6 Significance of the Study
The present study serves an important role to enrich the pool of knowledge of CAF
research. From the previous studies on CAF, it is worth noticing that there is no
conclusive result to explain how different planning conditions affect writers’ CAF,
except for the consistent findings which indicate pre-task planning has positive effects
on writers’ fluency (eg. Ellis & Yuan, 2004, Abdollahzadeh & Kashani, 2011, Johnson,
Univers
ity of
Mala
ya
12
Mercado & Acevedo, 2012, Piri, Barati & Ketabi, 2012 etc). However, the effects of
planning on accuracy and complexity are still a big area that needs to be explored.
Therefore, conducting a CAF research is a worthwhile effort to enrich the findings in
this area, with the hope of seeking for consistency with the findings of other studies.
Furthermore, to the researcher’s knowledge, there are very little studies of CAF
conducted in Malaysian context and even less was conducted on primary school
children. Hence, it is important to conduct a CAF research on Malaysian primary school
pupils’ writing to fill in the gap of the research in this area. With the findings obtained
from the present study, there will be more empirical data for second language
researchers to explain how Malaysian primary school pupils’ would actually perform, if
they are provided with the time to plan before performing a task or to write without time
pressure.
Furthermore, the present study investigates the viability of the Kellogg’s (1996)
Model of Process Writing hypotheses in the case of Malaysian Chinese primary school
pupils. According to Kellogg’s (1996) model, writers attend to different processes when
they write but due to the limited capacity of central executive system, writers will only
focus on a specific component of writing processes which they prioritise mentally. To
test the theory, the present study investigates whether pupils in Malaysian Chinese
primary school write as predicted by this model.
Moreover, the present study intends to bring some awareness to the writers and
English language teachers about the importance of planning, which is not emphasised in
the teaching of writing at primary school level. If planning has positive effects on the
writing quality, pupils in primary schools should be encouraged to plan before
performing their written task, even though in their examination, inflected keywords are
given according to the flow of the picture cues.
Univers
ity of
Mala
ya
13
Lastly, it is hoped that the present study would shed some light to improve the
practice of English language writing in the Malaysian context. With the comparison of
performing narrative writing task under pre-task planning, on-line planning and no
planning (time pressured) conditions, writers in the Malaysian context should adopt the
most favourable planning condition in their writing practice, so that they can write with
better quality.
1.7 Scope and Limitations of the Study
First of all, the limitation of the present study deals with the generalisability of the
results. Due to the unique multicultural background in Malaysia, pupils in Malaysian
Chinese schools need to learn at least three languages in schools (Ting, 2013).
Therefore, the challenge for the Chinese primary school pupils to learn second language
writing skills is different than other contexts. Therefore, the results of the research
might be more relevant to the unique multicultural context in Malaysia, even though
other cultural contexts might still be able to refer to these results in their second
language writing studies.
In addition, the study focuses only on the planning conditions research by Ellis
and Yuan (2004), i.e. pre-task planning, on-line planning and no planning. In previous
studies, other researchers used different planning conditions in carrying their studies.
For instance, Shin (2008) employed individual planning condition and collaborative
planned condition and Johnson, Mercado and Acevedo (2012) examined the sub-
processes of L2 writing, such as idea generation, organization and goal setting.
Furthermore, the present study investigates the subsystem of planning in
accordance to Kellogg’s (1996) Model of Process Writing. In this study, other systems
involved in writing are not prioritised in discussion unless the need of involving those
systems arises.
Univers
ity of
Mala
ya
14
Other than that, the present study investigates the effects of planning on the
quality of young learners’ writing. Due to the developmental differences between young
learners and adult learners, considerations should be made should there be any
transference of result or theory of young learners’ writing to adult learners’ writing and
vice versa.
Moreover, the researcher decided to research on narrative writing in the present
study due to the relevance of the genre to the researched context. Thus, the results
obtained from the study are limited to the studies of narrative writing, exclusive of other
genres of writing such as descriptive, argumentative, etc.
Next, the research is carried out to the children aged 12 in a Chinese primary
school. Pupils at other ages are not involved in the research due to the reason that they
are different in their language performance at different developmental stage.
1.8 Organisation of the Dissertation
The present study consists of six main chapters. The first chapter provides a brief
introduction regarding the background of the study, problems/ raised issues, purposes/
objectives of the study and etc.
The second chapter describes the literature related to the effects of planning on
the complexity, accuracy and fluency of L2 writers. This chapter introduces the
theoretical framework used in the present study, together with the rationale of the
selection of the theoretical framework.
The third chapter outlines the research design and methodology of the present
study. In this chapter, readers will be able to understand how the present study is carried
out in the setting, and how the collected data is analysed.
Chapter Four in the present study presents the results obtained from the study. In
this chapter, the detailed analysis of data obtained from each instrument is presented.
Univers
ity of
Mala
ya
15
The results in this study would be able to prove whether both planning conditions (pre-
task and on-line planning) affect CAF of Chinese primary school learners.
Chapter Five of the present study discusses about the results of the study.
Besides, this chapter gives justifications of the results yielded from the research. This
chapter also summarises the findings of the research.
Last but not least, the sixth chapter concludes the whole study and provides
suggestions and recommendation for future studies. It is a short chapter but it provides a
good insight for the future researchers to improve the current study.
1.9 Conclusion
This introduction chapter describes the background of the study, drawing from the
history of English language curriculum to the current practice of narrative writing in
Malaysian Chinese primary schools. Furthermore, it points out the problem statements,
purposes of the study, objectives of the study, research questions, the significance and
also the limitations and scope of the present study, which justify the needs of the present
study to be carried out.
Univers
ity of
Mala
ya
16
CHAPTER TWO: LITERATURE REVIEW
2.0 Introduction
This chapter discusses the important concepts and constructs related to the present study.
The essence drawn from the results, findings, justifications, predictions and relevant
information about the previous studies will also be included in this chapter.
2.1 Narrative Writing
In the present study, the genre investigated was the narrative writing of L2 young
learners. According to G. Ellis (2013), ‘young learners’ are children below 18 years old.
In the present study, the pupils selected were Year 6 pupils, who were 12 years old in
average. From studying young learners’ narrative writing, we would be able to identify
how young writers elaborate an event using their repertoire of writing skills.
A narrative is a story that describes a sequence of events (Moen, 2006). To
distinguish a narrative genre apart from other genres, there are some elements that a
reader can look into. Nevertheless, different notions and syntax categorisations of the
elements in a narrative were used by different scholars, for examples, studies by Labov
and Waletzky (1967) and Van Dijk (1977a). Labov and Waletzky (1967) identified
three macro-structures of a narrative: 1) Orientation, 2) Complication, 3) Evaluation.
Van Dijk (1977a) listed the superstructure of a narrative, which consists of five major
elements: 1) setting, 2) complication, 3) resolution, 4) evaluation and 5) moral. There is
also an alternative categorisation of the narrative superstructures, which a narrative
consists of setting and episode and the superstructure of episode can be further divided
into exposition, complication and resolution (Kintsch, 1976, as in Cortazzi, 2014).
Univers
ity of
Mala
ya
17
Figure 2.1. Model of Narrative Superstructure (Van Dijk, 1977a).
Figure 2.2. Alternative Model of Narrative Superstructure (Kintsch, 1976, as in
Cortazzi, 2014).
2.2 Second Language Writing
The current study falls under the niche of second language writing studies. Based on the
researcher’s experience in teaching Chinese primary schools, English language is their
second language. In the process of writing in English, pupils in Chinese primary schools
tend to translate words, phrases and organisation from their first language. According to
Connor (1996), translation from first language in ESL writing is very common even
among the advanced level students. In the translation process, the language and stylistic
convention from the writers’ native language and culture are borrowed (Connor, 1996).
Univers
ity of
Mala
ya
18
As the first language of the pupils in the present study are mostly Mandarin and
Hokkien, the instruction of ESL writing is somewhat different than teaching writing as a
first language. According to Silva (1993), L2 writing instructions should be different
from L1 writing in terms of their expectation in writing test as it is difficult to expect L2
writers to write as good as L1 writers in writing tests. Furthermore, L2 writing
instructors should be aware of learners’ sociocultural, rhetorical and linguistic
differences from the target language and they need to design the lessons to suit the
learners’ needs of learning. L2 writing teachers, unlike their L1 counterparts, need to
teach writing from the rhetorical, strategic and linguistic aspects as in Silva’s (1993)
study, L1 learners manifested greater sophistication in these three aspects. Therefore, it
is important for L2 writing instructors to teach learners planning, transcribing and
reviewing strategies which take rhetorical, strategic and linguistic aspects into account
(Silva, 19993). Hence, different perspectives, models and practices are often employed
in the instruction of ESL writing (Kroll, 1990).
Besides, the challenge of acquiring second language writing skills is even more
demanding when it comes to children. According to Lanauze and Snow (1989), L2
children (in their case, fourth and fifth graders in a Spanish-English bilingual
programme) need more time to the exposure of language input so that they can apply
their language skills properly in classrooms. To address this concern, the present study
seeks to find a planning condition which enables second language young writers to
maximise their performance in writing.
2.2.1 Measurement of L2 Performance: Complexity, Accuracy and Fluency (CAF)
Complexity, accuracy and fluency (CAF) are notions used in second language studies to
measure language performance (Pallotti, 2009). According to Gené-Gil, Juan-Garau and
Univers
ity of
Mala
ya
19
Salazar-Noguera (2015), CAF is objective indicators of the written performance of
language learners. CAF are used to evaluate oral and written performance because
researchers and language practitioners believe that they can adequately address the
multi-componential nature of L2 performance (Ellis and Barkhuizen 2005).
In the analysis of CAF in language production, language researchers have to be
very clear with the definition of CAF constructs. The construct of fluency is the
measure of real time language production at normal rate without hesitation or
interruption (Skehan, 2009). In terms of writing fluency, it is operationalised as the
aptitude of writers to produce huge chunks of texts and it is evaluated through using the
length of production units (Latif, 2012). It can also be regarded as how fast the writers
write in writing tasks (So, 2015). Nevertheless, the measure of fluency alone is
multifaceted in its nature. The measure of fluency can be categorised into three sub-
dimensions: 1) breakdown fluency, 2) speed fluency and 3) repair fluency. Breakdown
fluency is assessed by counting the pauses and amount of silence in language
production. Speed fluency can be evaluated based on speech and articulation rates,
mean length of run and time ratio. Repair fluency is measured by giving aspects such as
reformulation, replacement, false starts and repetition of words or phrases into
consideration (Tavakoli & Skehan, 2005). Nevertheless, researchers of writing gained
their insights primarily from the research of speech in measuring fluency (Waes &
Leijten, 2015). However in the studies of L2 writing, two of the fluency aspects which
have been measured are temporal fluency (rate of production) and repair fluency (Lin,
2013).
The construct of accuracy is another construct which is often used to measure
language performance. However, the use of this measure to indicate language
development has become a debateable question (Palloti, 2009). The construct of
Univers
ity of
Mala
ya
20
accuracy is defined as the ability of avoiding error in language production that might
reflect higher levels of language competency as well as the ability to avoid the use of
risky structures that might cause errors in production (Skehan & Foster, 1999).
According to Bley-Vroman (1983), the development of interlanguage cannot be judged
based on the accuracy measure per se. Despite the criticism of its validity, the measure
of accuracy is considered as a well-defined and coherent construct (Norris & Ortega,
2003: 737). Thus, measurement of accuracy should be assessed separately with the
construct of development (Pallotti, 2009).
Among the CAF measures, complexity is considered as the most problematic
construct (Pallotti, 2009). It is defined as the capacity of using more sophisticated
language (Skehan & Foster, 1999). According to Ellis and Barkhuizen (2005), there are
eight types of measures used in analysing learners’ language complexity in these five
aspects: 1) interactional (i.e. number of turns and mean turn length), 2) propositional (i.e.
number of idea units encoded), 3) functional (i.e. frequency of some specific language
function), 4) grammatical (i.e. amount of subordination, use of specific linguistic
features and mean number of verb argument) and 5) lexical (i.e. type-token ratio).
Nevertheless, in the field of task-based studies, Housen, Kuiken and Vedder (2012)
categorised linguistic complexity measures based on three major linguistic components:
1) grammatical complexity (i.e. syntactical and morphological complexity) and 2)
lexical complexity.
Moreover, measurement of CAF involves formulaic operationalisation of each
construct, which includes the calculation of ratios and frequencies. For instance,
syntactic complexity, accuracy and fluency are measured by counting number of
subordinations, number of errors and number of pauses, respectively (Norris & Ortega,
2009). Hence, Skehan (2009) listed the descriptions of a well performed language task,
Univers
ity of
Mala
ya
21
which includes more advanced language, error avoidance and uninterrupted rate of
language production.
In the application of second language acquisition (SLA) research, the use of
CAF as a measurement is no longer a novel approach. The use of CAF in SLA studies is
to explain the development of language competency of learners in response to tasks,
stimuli or teaching (Norris & Ortega, 2009). The present study is extended from the
numerous studies investigated on the effects of different task conditions to the language
quality of language learners but the results are still inconsistent. CAF measures are
chosen in the present study because of their multidimensionality. It is considered
multidimensional as in Housen and Kuiken (2009), they stated that CAF has been used
to assess different types of language domains using various forms of tools, ranging from
subjective and holistic ratings by using raters to qualitative and more objective
measures such as frequencies, ratios and formulas. Due to their multidimensionality,
CAF measures are chosen to evaluate learners’ language performance in a holistic and
precise manner. In the present study, the researcher only employed objective measures
(such as syllables per minute, percentages of error-free clauses and ratio of clauses to T-
units) to evaluate CAF. Nevertheless, these measures cover all three aspects of CAF.
Other than that, these measurements are used in the present study because they have
been used in L2 writing before. The results obtained in this study can be compared to
other previous studies and from the comparison and hence, a more subjective
explanation of the effects of planning on young learners’ CAF can be given.
2.3 Planning in Second Language Writing
Planning is a process that forms mental pictures of knowledge to be applied in writing.
Most of the time, it is greatly different than the written product in terms of the degree of
Univers
ity of
Mala
ya
22
abstraction (Flower and Hayes, 1981). Flower and Hayes (1981) also pointed out the
subprocesses involved in the planning process of writing, as shown below:
a) Generating Ideas: The retrieval of information from long-term memory.
b) Organising: The process of structuring meaning to the writer’s ideas.
c) Goal-setting: The process of establishing a purpose and short-term goals in
writing (Hacker, Dunlosky& Graesser, 1998). It is also a process created by the
writer which goes on throughout the process of composing. It is also the major
process in planning and it plays as an important part to determine the
creativeness of writers, or even to determine good and bad writers.
Nevertheless, previous studies have shown that young learners do not plan their
writing properly. According to MacArthur, Harris and Graham (1994) in Berninger et.
al (1996), younger writers are less likely using strategies such as using text structure
knowledge, employing heuristic strategies to search for content from the memory or
directing the writing process using goals. Moreover, Burtis, Bereiter, Scardamalia and
Tetroe (1983) in Berninger et. al (1996) also believe that young writers have difficulty
to separate on-line planning with translation process, whereas the more experienced
writers acquired the ability to perform ‘advanced planning’ before and separate from the
text production. In relation to this, a study conducted by Graham, Harris, MacArthur
and Schwartz (1991) reveals that both average and learning disabled young learners
(aged between 9 and 12) in primary schools spend less time in such form of ‘advanced
planning’ but they spend more time in on-line planning and translating. From the
literature, the researcher interpreted ‘advanced planning’ as pre-task planning as in
Berninger et al’s (1996) study, writers who underwent advanced planning were given 10
minutes to plan before performing the task. The operationalisation of advanced planning
used in Berninger et al’s (1996) is similar to the operationalisation of pre-task planning
used in the present study.
Univers
ity of
Mala
ya
23
The process of planning, however, can occur before the writers perform the task
and/ or while the writers are performing the task. From the literature to date, there are
two types of planning identified in L2 writing: pre-task planning (PTP) and on-line
planning (OLP) (Hsu, 2012). The first type of planning, which is PTP can be further
divided into rehearsal and strategic planning (Ellis, 2005). Rehearsal is the repetition of
task prior to the actual writing (Hsu, 2012) whilst strategic planning is the preparation
done for a language task by considering the encoding and expression of the actual task
materials (Ellis, 2005).
In contrast with PTP, OLP is the planning process that occurs within the task
and operationalised as the providence of as much time as the writers need in a written
task (Ellis, 2009). With the on-line planning, writers will have less pressure in
communication (Skehan and Foster, 2005).
In the present study, the effects of both PTP and OLP are compared in order to
investigate whether these two planning conditions can really yield better quality of L2
writing. Besides, the baseline performance of the young L2 writers is obtained in the
present study by having a control group, which L2 writers perform the writing task
without performing any means of planning. Having this control group is a step to
examine the effectiveness of planning (be it PTP or OLP) in improving the quality of L2
learners’ writing.
2.3.1 Planning in L2 Writing: Theoretical Framework
Writing is an activity which involves several mental processes. There are various
theories or models, such as those proposed by Flower and Hayes (1981), Bereither and
Scardamalia (1987), Grabe and Kaplan (1996), Kellogg (1996) and Zimmerman (2000)
Univers
ity of
Mala
ya
24
that explain the mental processes involved in writing and most of the theories are
similar in many ways. In the present study, Kellogg’s (1996) Model of Writing
Processes is chosen as its theoretical framework. Even though it was previously used to
describe L1 writing, Kellogg’s (1996) model has been extensively used in studies L2
writing to interpret the results of how planning affect writing production (Ellis & Yuan,
2004; Johnson, Mercado, Acevedo, 2012; Seifollahi & Tamjid, 2012).
Kellogg’s (1996) model is chosen for the present study because the model
describes the systems of writing and the interaction of them in details (Cho, 2003).
Kellogg’s (1996) model posits that there are three main systems involved in writing: 1)
Formulation, 2) Execution and 3) Monitoring (Ellis & Yuan, 2004). In each system, two
writing processes are involved and those processes are linked in a sequential order.
Furthermore, the model is used because it can find out the correct location of the
different writing processes in each of the working memory registers (Alamargot &
Chanquoy, 2001). Other than that, Kellogg’s (1996) model is the only model which
provides the link between the planning subprocess and its relating working memory
components. Based on these reasons, the model is chosen to be the theoretical
framework in the present study.
The two processes involved in the formulation system are planning, and
translating (Kellogg, 1996). Process of planning entails goal establishment, idea
creation and the organisation of these subprocesses to facilitate action (Ellis & Yuan,
2004). The process of translating is the process which relevant lexical units and
syntactic frames are selected so that these linguistic units can be encoded and ready to
be represented either in graphological and phonological forms in the execution system
(Ellis & Yuan, 2004). To further describe the process of planning, a writer plans his
writing by visualising a few things, such as ideas, supporting graphics, organisational
Univers
ity of
Mala
ya
25
schemes, layouts and orthography output (Kellogg, 1996). Then, the engagement of the
visuo-spatial sketchpad enables the writer to conjure visual imagery of ideas which is
drawn from the long-term memory (Paivio, 1986; as in Kellogg, 1996). Hence, the
writing literature always regards planning as the non-linguistic work. The subprocess of
translating, which is regarded as the linguistic work from literature, is the process
involved selection of lexical units, building of syntactic frame, phonological
representation of lexical units in frame and transference of phonemes into graphemes.
However, it should be noted that planning and partial translation process are commonly
happened concurrently (Kellogg, 1996). Thus, when a writer performs both planning
and translating at the same time, both processes compete for the working memory
resources and therefore the competition leads to the increase of the cognitive demand
(Johnson, Mercado, & Acevedo, 2012).
In the execution system, the output from the translation subprocess will be
programmed by handwriting, typing (Abdel Latif & Muhammad, 2009) or dictating
(Kellogg, 1996). The two subprocesses involved in this system are programming and
executing. Programming deals with the motoric elaboration whilst executing is the
realisation of the message (Alamargot & Chanquoy, 2001). However, the subprocesses
of programming and muscle execution differ based on their output modes (handwriting,
typing or dictating). In handwriting for example, a writer has to select the appropriate
letter size and allographic form, setting the force parameter and get the muscles ready
for the writing process (Van Galen, 1990 in Kellogg, 1996). In the execution subprocess,
a writer needs to execute processes such as motor commands, muscle movements and
feedback mechanisms, which ensure the legibility of the handwriting (Shepard, 1994;
Brown 1988 in Kellogg, 1996).
Univers
ity of
Mala
ya
26
The final system which a writer would engage cognitively is the monitoring
system. The two subprocesses involved in the system include reading and editing. In
the monitoring system, the language output produced by the writer is evaluated and
revised (Becker, 2006). In the subprocess of reading, the central executive and the
phonological loop are activated as the writer reads the text he produced while in the
editing subprocess, visuo-spatial sketchpad is involved in maintaining the correct
picture of how the text is placed in a page (Galbraith, 2009).
To relate Kellogg’s (1996) model to the present study, this model is chosen as
the theoretical framework due to its central claim that in the production of text, a writer
has to decide which subprocess of writing should be prioritised due to the fact that the
central executive does not have the capacity to attend all subprocesses of writing (Ellis,
2005). Nevertheless, the three systems involved in Kellogg’s model should not be
understood as a one-after-another process. Instead, these three distinct systems operate
simultaneously when a writer writes a text (Kellogg, 1996). Therefore, writing is a
mentally complex process which various subprocesses are involved concurrently.
Getting to know the nature of the intrinsic part of the writing process might help SLA
researchers to identify the manipulation of the subprocesses involved in writing and
therefore maximise learners’ writing performance. In the next sections, the roles of
working memory will be presented, as it is an important component in the process of
writing. An illustration of the model is shown in Figure 2.3. Univers
ity of
Mala
ya
27
Figure 2.3. Kellogg’s (1996) Model.
Other than Kellogg’s (1996) model, there are other models which illustrate the
process of writing, such as Flower and Hayes’ (1981), Bereither and Scardamalia’s
(1987), Grabe and Kaplan’s (1996) and Zimmerman’s (2000) models as presented in
Figure 2.4, Figure 2.5, Figure 2.6 and Figure 2.7. The other four models are presented in
this section in order to compare their suitability to be used in the present study.
Univers
ity of
Mala
ya
28
In Flower and Hayes’ (1981) model of writing (demonstrated in Figure 2.4), the
three main components involved in writing are the task environment, the long-term
memory of the writer and the writing process (Flower and Hayes, 1981). According to
the model, the task environment contains all of the external elements, ranging from the
rhetorical problem or assignment and the text produced at the moment. The second
component, the long term memory, dealt with the writer’s knowledge of the topic as
well as the variety of writing plans and audience. The third component of the model is
the writing processes, which are planning, translating and reviewing as illustrated in the
model. The three processes of writing, according to the model, are controlled by a
monitor (Flower and Hayes, 1981). Nevertheless, this model is not suitable for the
present study as the main focus in the present study is on the writing process but not on
the task environment and writer’s long term memory.
Figure 2.4. Flower and Hayes’ (1981) Model.
Moreover, Bereiter and Scardamalia’s (1987) knowledge-transforming model
(illustrated in Figure 2.5) posits that content generation in composition involved the
interaction of two problem spaces: 1) a content space which problems of belief,
Univers
ity of
Mala
ya
29
consistency of logical thought and the like are processed and 2) a rhetorical space where
problems pertaining to the composition goals are handled (Bereiter & Scardamalia,
1987). In developing a main point in composition, a problem-solving process which
involves the interaction between content problems and rhetorical problems is taken
place (Bereiter & Scardamalia, 1987). However, this model does not include the role of
working memory in writing, which is one of the important elements used in the
discussion of findings in the present study. Hence, this model is not chosen to be the
theoretical framework of the present study.
Figure 2.5. Bereiter & Scardamalia’s (1987) Knowledge Transforming Model of Writing.
In the attempt to take the cognitive processing of writers, the linguistic and
textual resources and contextual factors into account, Grabe and Kaplan (1996)
postulated a model of writing for communicative language use (Lu, 2010). The model
Univers
ity of
Mala
ya
30
(as shown in Figure 2.6) includes two important components, i.e. context and verbal
working memory in the model. ‘Context’, which is the external factor, consists of
situation and performance. ‘Situation’ refers to the participants, setting, task, text and
topic involved in the text production. ‘Performance’, which is the actual produced text,
refers to the product resulting to the processing of verbal working memory. In Grabe
and Kaplan’s (1996) model, verbal working memory comprises three subcomponents: 1)
internal goal setting, 2) verbal processing and 3) internal processing output. ‘Internal
goal setting’ plays its role to mediate the external factors from the context to ‘verbal
processing’ (which draws on language competence, knowledge of the world and on-line
processing assembly as illustrated in Figure 2.6). The outcome of verbal processing is
‘internal processing output’ which is a processing output that externalises language
performance in forms of textual output (Lu, 2010). From the description of the model, it
is apparent that the model seeks to relate production of writing to its social context and
cognitive processing for communicative language use. Nevertheless, this model is not
suitable to be adapted in the present study as the present study needs a model which
describes the process of planning in writing.
Univers
ity of
Mala
ya
31
Figure 2.6. Model of Writing as Communicative Language Use (Grabe and Kaplan, 1996)
Drawing on the social cognitive theory, Zimmerman (2000) postulated a cyclical
model of self-regulated learning (see Figure 2.7). Based on the model, learners would
go through three self-regulatory phases in learning, i.e. 1) forethought, 2) performance
and 3) self-reflection. The forethought phase consists of two areas, which are task
analysis and self-motivation beliefs that prepare learners to learn and enhance that
learning. In the performance phase, processes such as self-control and self-observation
are employed to facilitate learners’ performance. In the self-reflection phase, learners
react to their learning outcomes and influence the forethought phase in the coming cycle
(Zimmerman, 2013). The model proposed by Zimmerman (2003) does not meet the
need of the present study as it describes the cyclical learning process rather than the
roles of planning in the language production.
Univers
ity of
Mala
ya
32
Figure 2.7. Zimmerman’s (2000) Self-regulated Learning Model.
Based on the comparison of Flower and Hayes’ (1981), Bereiter and
Scardamalia’s (1987), Grabe and Kaplan’s (1996) and Zimmerman’s (2000) models
with Kellogg’s (1996) model, it is apparent that Kellogg’s model is the most suitable
model to be used as the theoretical framework in the present study, as it illustrates the
processes involved in the production of L2 writing as well as the roles of working
memory in the process of planning and other processes involved in writing. Hence, from
the comparison of the existing models in L2 writing, Kellogg’s (1996) model is selected
to be used in the present study.
2.3.2 Working Memory in Writing
As it has been known, writing is a complex cognitive process. Thus, the process of
writing involves the use of working memory. Kellogg’s (1996) model is adopted from
Baddeley’s multicomponent model of working memory to work on language production
Univers
ity of
Mala
ya
33
especially for writing (Olive, 2004). According to Baddeley (1992), there are three
subcomponents of working memory: 1) central executive, 2) visuo-spatial sketchpad
and 3) phonological loop.
In general, the investigation of working memory in writing can be researched in
two different points of view. First, the roles of working memory can be investigated by
studying the demands of writing tasks on short-term memory which specifically
examines the relationships between writing processes and slave systems which are the
phonologicial loop and visuo-spatial sketchpad according to Galbraith (2009). Second,
the roles of working memory can also be studied by looking at how the central
executive manages its functions and directs its attention in writing processes (Olive,
2004). In the next sections, the roles of visuo-spatial sketchpad and central executive
will be discussed because these two components are related to the process of planning,
which is one of the writing processes investigated in this study.
2.3.3 Roles of Visuo-spatial Sketchpad (VSSP) in Writing
As mentioned in the previous section, visuo-spatial sketchpad is one of the components
involved in the process of planning. Visuo-spatial sketchpad is one of the two
subsidiary systems in working memory (Baddeley, 2004). However, this component of
working memory has received less attention for research in comparison to the other
component in the working memory, the phonological loop (Logie, 1995).
Visuo-spatial sketchpad plays its role to generate, manipulate and retain visual
images (Stainthorp, 2004). In another way, it maintains the visual representations and
the spatial representations in the writing process (Schacter, Wagner & Buckner, 2000).
Olive and Passerault (2012) state that the engagement of the visuo-spatial working
Univers
ity of
Mala
ya
34
memory allows a writer to select and organise information to be included in texts.
Besides Olive and Passerault (2012) also state that a writer can search for ideas, process
and integrate pictures and retrieve conceptual knowledge from his long-term memory
with the engagement of the visuo-spatial sketchpad. Hence, visuo-spatial sketchpad
plays a vital role in retrieving and structuring information needed in writing.
Olive and Passerault’s (2012) statement is in line with the assumptions made by
Kellogg (1999) which assumed that the VSSP is activated when a writer processes
figurative materials. The spatial element of the VSSP is engaged when a writer
organises information (Olive, Kellogg & Piolat, 2008). In the process of planning in
particular, many researchers suggested that visuo-spatial components of working
memory are engaged during text production (Olive, Kellogg & Polat, 2008). Therefore,
it is evident that the VSSP is an important component involved in the writing process,
especially during the planning subprocess which requires the retrieval of image-based
conceptual content (Olive, 2004). In other words, writers withdraw the visualised
images from VSSP in the planning subprocess and in the translating subprocess, verbal
components are drawn from the VSSP in order to formulate the lexical units and
syntactic frames needed in writing (Yi & Luo, 2013).
2.3.4 Roles of Central Executive Unit in Writing
According to Baddeley and Logie (1999), the central executive unit controls and
regulates the working memory system. They also listed its functions as follows:
a) Coordinating the phonological loop and VSSP,
b) Manipulating attention in terms of focusing and switching and
c) Engaging representations within long-term memory.
Univers
ity of
Mala
ya
35
Other than the aforementioned functions, Miyake et al. (2000) described
three general functions served by the central executive system:
a) Inhibition: refers to the ability to consciously inhibit automatic, dominant
and prepotent responses when necessary.
b) Updating: with this function, our brain will monitor and code information
and link it to the task and at the same time, replace the old or no longer relevant
information accordingly.
c) Shifting: a function concerning alternation of attention between tasks,
operations or mental sets.
To relate the functions of the central executive unit in writing, it is important for
SLA researchers to know how the central executive unit plays its roles in directing
writers’ attention in writing. In the present study, the researcher would like to
investigate how writers attend to the note expansion task and at the same time to explain
the phenomenon based on theories of the manipulation of central executive unit
functions.
2.3.5 Roles of the Phonological Loop in Writing
The phonological loop is a storage system which stores speech-based information and
acoustic information (Henry, 2011). The two subcomponents in the phonological loop
are the phonological store and the articulatory rehearsal mechanism. The phonological
store keeps the materials used for writing temporarily, whereas the articulatory rehearsal
mechanism plays its role to rehearse the temporary information from the phonological
store to prevent loss of information (Henry, 2011). To relate the phonological loop in
Kellogg’s (1996) model, it plays a role for learners to read the text they produced and
edit it when necessary.
Univers
ity of
Mala
ya
36
2.3.6 Assumptions in Human Information Processing and Their Links to
Kellogg’s (1996) Model
Writing is a highly complex activity which involves different processes simultaneously.
These processes have been identified in different models of writing, for instance the
Kellogg’s (1996) model which is selected for the present study. Moreover, writing is
also known as a human information processing activity (Stainthorp, 2004). In the effort
of explaining writing process using Kellogg’s (1996) model of working memory, there
are several assumptions that writing researchers need to make in order to make the
theory plausible. First of all, one of the assumptions and also the key feature of
Kellogg’s (1996) model is the claim of limited capacity. In Kellogg’s (1996) model, the
central executive is seemed to have limited capacity and when a writer performs a task
under pressure, s/he needs to decide which writing process s/he needs to prioritise.
Consequently, a trade off of attention would happen for a writer to dedicate attention to
particular writing processes. Generally, formulation process seems to be more
cognitively demanding and hence, it takes priority over execution and monitoring
processes (Ellis, 2005).
Apart from that, the second assumption which relates the activity of writing to
human information processing is the claim of control mechanism. According to
Stainthorp (2004), a writer needs to have a control mechanism to perform writing tasks.
This mechanism controls the processing, storage, retrieval and utilisation of information
needed in writing tasks. Thus, a writer’s capacity in performing writing tasks is always
dedicated to different aspects of the executive function with different proportions.
When a writer is performing a new writing task, more capacity will be dedicated to the
compositional aspects of the task rather than the routine and overlearned aspects.
Univers
ity of
Mala
ya
37
Therefore, a well trained writer may need less effort in writing because very little
capacity is required for him to perform the non-compositional aspects of writing, such
as spelling the words and synchronising motor movements.
Furthermore, in accordance to Kellogg’s (1996) model, the differences of
allocation of mental resources between pre-task and on-line planning exist as the central
executive system has limited capacity so prioritisation of form or meaning would take
place in different planning conditions. Other than that, due to the limited attentional
capacity of second language learners, they experience trade-off effects which cause the
competition between complexity, accuracy and fluency of language production (Skehan
& Foster, 1997). Hence, the present study investigates how young writers prioritise
these components in different planning conditions. In Skehan and Foster’s (1997) study,
they discovered that there is an interaction between task type and planning. Learners
produced more accurate speech when they were given structured task with the
opportunity to plan and they tended to produce speech with greater complexity when the
tasks provide them the opportunity to attend on-line planning or require complex
outcomes. The present study examines how young ESL writers allocate their mental
resources under different planning conditions and the effects of the planning conditions
on their language production.
2.3.7 Reasons of Selecting Kellogg’s (1996) Model as Theoretical Framework
Kellogg’s (1996) model is chosen for the present study because this model clearly
describes the components involved in a writing process. Besides, it provides SLA
researchers the basis of how planning strategies affect language production (Ellis, 2005).
For instance, with the availability of pre-task planning, a writer could reduce his
cognitive loads of his working memory resources and therefore able to put greater
Univers
ity of
Mala
ya
38
attention on the translation process when he writes (Ellis & Yuan, 2004). On the other
hand, on-line planning enables a writer to monitor internally processed language in the
translation process before externalising the actual textual output (Ellis & Yuan, 2004).
Therefore, Kellogg’s (1996) model provides SLA researchers solid descriptions of how
writers attend to writing tasks under different planning strategies.
Other than that, using Kellogg’s model as a theoretical framework provides us
with the descriptions of L2 writing processes and therefore, makes it easier for L2
writing researchers to hypothesise the effects of planning on learners’ performance
(Ellis, 2005). According to Kellogg’s (1996) model, learners who perform pre-task
planning can write more fluently because pre-task planning reduces the cognitive load
of working memory and writers can fully focus on the translation process as they write.
Furthermore, Kellogg’s (1996) model also suggests that the use of complex
grammatical forms can be increased as the demands of working memory resources have
been reduced. Nevertheless, there is no denite prediction made by this model in regards
to the effects of planning on the complexity of writers’ production (Johnson et al., 2012).
Apart from that, Kellogg’s (1996) model is chosen in the present study because
it is closely related to working memory. Different than other models such as the model
of writing by Grabe and Kaplan (1996) and Flower and Hayes’ model (1981), this
model relates the writing processes with writing system and connects them with
relevant working memory components. The descriptions of the model help L2
researchers to identify a more accurate location of the different writing processes in
each of the working memory registers (Alamargot & Chanquoy, 2001).
Univers
ity of
Mala
ya
39
2.4 Relevant studies on L2 Language Production Performance
In terms of linguistic studies, the research of planning effects on language production
could be traced back to late 1980s. From the available literature, the first relevant study
was carried out by Ellis (1987). In his study, he examined the effects of planning on the
use of past tense markings in written and oral narratives. The three structures which he
investigated were 1) regular past tense form, 2) irregular past tense form and 3) copula
past tense form. To analyse the data, he counted the percentage of correct and incorrect
verb forms used in the learners’ written and spoken discourse. From his research, he
discovered that L2 learners tend to produce more target like of accuracy when they were
given the opportunity to plan. And to compare with two modalities of production,
writing seemed to have more advantage in gaining higher percentage of language
accuracy compared to speech.
Adding in another dimension of analysis, Crookes (1989) investigated on how
planning affects the oral interlanguage of 40 Japanese students. Apart from looking at
the language accuracy, he included the measures of language complexity in his study.
There were 11 measures of language accuracy and complexity altogether in his study
which included VP range per utterance, type per token ratio, words per error free and
etc. The students involved in his study performed the oral task in two conditions, which
are with planning and without planning. In terms of complexity measures, the study
concluded that learners produced more complex interlanguage when they were given
the opportunity to plan. Besides yielding more complex language, the study also
suggested that planning might have some slight effects in improving learners’ language
accuracy. Other than that, planners were also found to use more discourse markers
compared to non-planners in his study.
Univers
ity of
Mala
ya
40
Following the trend of Crookes (1989), Williams (1992) conducted a study
about the effects of planning on the use of discourse marking of 24 nonnative speaking
teaching assistants’ discourse. In her research, she also measured the comprehensibility
of the discourse produced by the participants using a variety of methods, such as using
raters to rate the discourse according to a scale similar to Speaking Proficiency English
Assessment kit (SPEAK), post rating interviews for the raters and also posing some
comprehension questions for the raters. Her study also sought to find a link between the
use of discourse marking and the comprehensibility of the discourse. The results of her
study showed that when the participants were in the planned condition, their
productions tend to be marked overtly and explicitly with discourse marking. Other than
that, she also discovered that the increased use of discourse marker improved the
comprehensibility of the spoken discourse as well.
The very first study which researched on the planning effects in terms of
language fluency, accuracy and complexity was done by Foster and Skehan (1996). In
addition, the participants in the study performed the given oral tasks (personal
information exchange, narrative and decision-making) under three planning conditions
(unplanned, planned without detail and detailed planning). From their study, they
discovered that the degree of planning directly influenced the degree of complexity. The
study also revealed that the effects of planning on fluency were complicated as from the
study, planning seemed to yield more repetition, hesitation and replacement.
Furthermore, mixed results were obtained in the accuracy measures, showing that
different task types might affect the effectiveness of planning on language accuracy.
Despite the mixed and unexpected results obtained from the fluency and accuracy
measures, this study had clearly gained more attention from applied linguists to research
on the connection between planning, task-type and language performance. Even though
the planning conditions employed in Foster and Skehan’s (1996) study were different
Univers
ity of
Mala
ya
41
from those used in the present study, their study provide the researcher of the present
study some insights about how planning would affect the performance of the learners
and how planning can be done in the time provided.
In the following year, Wigglesworth (1997) carried out another study focusing
on the oral fluency, accuracy and complexity measures. This study examined planning
effects based on the aforementioned three measures which involved 120 students with
different levels of proficiency. The study revealed the relationship between learners’
level of language proficiency and the effectiveness of planning on the different language
constructs, suggesting that the use of planning might only be beneficial to learners with
certain level of language proficiency.
Moreover, the research of planning on language production had also ventured
into the investigation of durational effects. For instance, Mehnert (1998) investigated 31
learners of German to perform oral tasks under different planning time: 0, 1, 5 and 10
minutes of planning. From the research, Mehnert (1998) found that fluency and lexical
density of speech increase as a function of planning time. In terms of accuracy, the
researcher found that planning had had some positive impact in improving the language
production but the extra provision of planning time does not further improve the degree
of language accuracy. For the measures of complexity, the researcher discovered that
planning would only benefit the planners if they were given more time to plan (as
suggested by Mehnert’s study, more than 10 minutes). Little time for planning did not
bring significant effects on the complexity of language production. Even though the
findings of the research were not conclusive, the research had opened another
possibility for future research to investigate the effects of planning time on language
performance.
Univers
ity of
Mala
ya
42
From the research presented from the 1980s to 1990s, it is apparent that most of
the research reviewed in that era was on oral tasks. The rationale of selecting these
studies to be reviewed is to present the need of conducting relevant research on written
tasks, as suggested by Soleimani and Kargozari (2014).
Other than judging the language productions in linguistic measures such as
complexity, accuracy and fluency, studies on the effects of planning are also extended
to other measures using analytical ratings. For instance, Shi (1998) used Hamp-Lyons’
scale to rate 125 ESL essays in order to examine the effects of pre-writing activities
such as peer discussion and teacher-led discussion on the quality of writing. The rating
scale used by Shi (1998) consisted of analytical measures such as global,
communication, organisation, argumentation, appropriacy and accuracy. From the
analysis, Shi (1998) found no significant difference of the quality of writing across the
three conditions of prewriting activities (peer discussion, teacher-led discussion and no
discussion). Instead, Shi (1998) found significant differences on the students’ length of
essays and use of vocabulary across groups, showing that students write longer when
they do not discuss about the essay neither with peers nor the teacher but they use more
verbs when they have discussion before writing. Nevertheless, the study of Shi (1998)
gives us more possibilities about how learners’ language production can be analysed for
relevant studies.
From the aforementioned research carried out from the late 1980s to 1998, the
studies of the effects of planning generally investigated the impact of planning on
language development. However, very few studies were done to explain the cognitive
processes during planning. In this much unexplored area, the study by Ortega (1999)
was considered as one of the earliest works which connected theories of cognitive load
to the planning process. Ortega’s (1999) study discovered that pre-task planning can
Univers
ity of
Mala
ya
43
increase learners’ focus on form in L2 oral productions. In explaining the finding,
Ortega (1999) claimed that planning reduces the cognitive load of L2 learners and
allows learners to give more attention on assessing task demands. Hence, learners will
be focusing more on forms when they are given the opportunity to plan their speech
prior to the oral task performance.
Since the introduction of cognitive load theories in language production studies,
there was (and still is) a growing concern of linking task structure and processing load
of language tasks in SLA studies. For example, Skehan and Foster (1999) conducted a
study which linked the inherent task structure of narrative oral task to the concept of
processing load. From their study, they discovered that fluency was predominantly
influenced by the task structure and complexity was influenced by the cognitive
processing load. However, their study found a complex result for the accuracy measure
which they explained the accuracy of performance was affected by the interaction of
task structure and processing load. Skehan and Foster’s (1999) had inevitably invited
more studies to explain the interaction of these two factors with language production.
From the studies conducted since the late 1980s to 1990s, it is easy to notice that
most of the studies focused on oral productions. Besides, previous literature showed an
increase of research which focused on writing after 2000s. Ellis and Yuan (2004)
carried out a study to investigate the effects of three different planning conditions (pre-
task planning, on-line planning and no planning) on 42 Chinese undergraduate English
major students’ fluency, accuracy and complexity in narrative writing. In their study,
they used a narrative writing task which consisted of six pictures. To compare with all
three planning conditions, Ellis and Yuan (2004) discovered that pre-task planning is
the best planning condition to improve students’ written fluency and on-line planning is
the best planning condition in yielding the most accurate written production. In terms of
Univers
ity of
Mala
ya
44
language complexity, the results obtained from their study generally showed that
planners do have more advantage over non-planner in producing written narratives
(except for the measure of lexical variety which showed no difference across the three
groups). This study is also one of the primary references for the present study given the
research design and methodology were well structured and well operationalised.
Similar to the research conducted by Shi (1998), Shin (2008) conducted a study
of how individual and collaborative planning affect the quality of 157 Korean learners’
essays. In the study, learners were required to write in both argumentative and
expository writing. Even though the study showed no significant mean differences
between both planning, the study found that Korean EFL writers’ written quality was
influenced by planning conditions and proficiency levels.
Relevant studies on second language writing have been carried out in recent
years, with slight difference in terms of research design and methodology.
Abdollahzadeh and Kashani (2011) researched on the 107 Iranian EFL learners
regarding the effects of task complexity on learners’ CAF in narrative writing. In their
study, they divided the learners into high and low proficiency groups and each of the
groups was further divided into Here and Now (low complexity task) and There and
Then (high complexity task) groups. Their study revealed that the effects of task
complexity towards language complexity and accuracy were significant but not on
language fluency, indicating learners produced writing with better accuracy and
complexity in the complex task. Furthermore, their research also showed that learners
with higher proficiency level were more likely to benefit from the manipulation of task
complexity. This study has given an insight to the researcher to be mindful with the
complexity of the task used in the present research.
Univers
ity of
Mala
ya
45
Other than that, Piri, Barati, and Ketabi (2012) tried to examine the effects of
combining both pre-task planning and on-line planning on the writing on 45 Iranian
learners of English. From their study, they discovered that the pre-task planning group
outperformed the on-line planning group significantly but did not outperform the
combined planning condition group in terms of fluency. However, no significant
differences across group were found for the accuracy and complexity measures.
Nevertheless, the effects of combining both pre-task planning and on-line planning still
need to be examined in future research.
Other than combining planning conditions, Salimi, Parviz, and Hosseini (2012)
also tried to combine planning conditions and task complexity in the language accuracy
of 50 English language learners from an English institute in Iran. In their research, they
found that strategic planning is beneficial in improving learners’ language accuracy,
regardless in simple or complex tasks. Their research finding was contradictory with
Piri, Barati and Ketabi’s (2012) and hence, it showed that there is an inconsistency of
results obtained from different studies. To explain this inconsistency, more studies need
to be carried out in order to prove the effectiveness of strategic planning in L2 language
accuracy.
Amongst all the studies conducted in this area, one of the studies which involved
the most number of participants was the research carried out by Johnson et.al (2012). In
their study, 968 learners of EFL were assigned into different pre-task planning groups,
investigating the effects of pre-task planning (idea generation, organisation, goal setting
and a condition with both goal setting and organisation) on learners’ language fluency,
grammatical complexity and lexical complexity in argumentative writing. From their
study, they revealed pre-task planning yields different effects in L2 writing than L1
writing because in their study, pre-task planning did not show greater effects on both
Univers
ity of
Mala
ya
46
lexical and grammatical complexity. Nevertheless, they claimed that pre-task planning
might have some positive effects in L2 writing. However, the potential effects were too
small to be considered significant. Johnson et. al (2012) attributed the ineffectiveness of
pre-task planning in their study to the proficiency level of the learners, claiming that
pre-task planning would only benefit learners with certain proficiency level of English
language. From the surprising findings obtained in their research, the effectiveness of
pre-task planning still needs to be justified by conducting more research.
Focusing only on language accuracy, Bagheridoust and Fakoor (2013)
conducted a research to investigate how pre-task and on-line planning affect the
performance of argumentative writing among 24 learners. Their study indicated that
pre-task planning yielded a positive effect in improving learners’ accuracy significantly
in terms of the percentage of error-free clauses produced in texts. On-line planning
seemed to be the best planning condition to enhance accuracy of percentage of correct
verb form but there was no statistical significant difference found in comparison to
other planning conditions. However, the findings obtained from their study could not be
a good representative findings given that the research was conducted using a very small
number of samples.
The trend of studying effects of planning in L2 written CAF still continued and
from the study carried out by Ghavamnia, Tavakoli, and Esteki (2013), the results were
quite similar to the results obtained from Ellis and Yuan’s (2004) study. In Ghavamnia,
Tavakoli and Esteki’s (2013) study, they found that pre-task planning could bring
beneficial effects in improving written fluency whereas on-line planning could improve
learners’ written accuracy. However, their study found that syntactic complexity is
influenced more by pre-task planning because from their study, the learners in the pre-
task planning group produced more clauses in comparison to its counterpart. Their study
Univers
ity of
Mala
ya
47
had given confirmation of Ellis and Yuan’s (2004) study but still the reasons behind the
effects of different types of planning on different constructs of language need to be
investigated.
A similar study which studied the effects of pre-task and on-line planning on L2
writing was also conducted by Lin (2013). The study discovered that neither pre-task
nor on-line planning had any effect on the 75 non-English major freshmen’s accuracy
and syntactic complexity of compare/contrast type of essays. In Lin’s (2013) study, it
was also discovered that pre-task planning had a beneficial effect on lexical complexity
and repair fluency. The study also found that both planning conditions had a negative
effect in learners’ speed of writing.
Other than this, the research conducted by Abdollahzade and Taak (2014)
examined the effects of planning time in the quality of writing which was measured
analytically based on content, organisation, vocabulary, language use and mechanics. In
their study, the planning time was provided with different lengths (no planning time, 5
minutes and 10 minutes). The results of their study indicated that planners had better
advantage than non-planners in writing and the planners who had the longest duration
for planning (10 minutes in their study) performed the best in expository writing. Their
research has provided an insight to other researchers and language teacher to allocate
appropriate duration of time for the learners to plan prior to writing in order to obtain
the best writing performance from the students.
To date and to the researcher’s knowledge, the most recent research carried out
in relation to planning conditions in writing is the study carried out by Yi and Ni (2015).
Their study investigated the effects of pre-task planning and on-line planning on the
syntactic complexity, lexical complexity, accuracy and fluency of 31 university students’
argumentative writing. The results of their study showed that pre-task planning had a
Univers
ity of
Mala
ya
48
positive effect on lexical complexity whereas on-line planning enabled writers to write
more fluently. Moreover, their study indicated that both planning conditions had no
significant effect on accuracy and syntactic complexity.
Besides the aforementioned research described in a chronological order, studies
of planning on the quality of writing have been carried out in Malaysia in recent years.
For instance, Sim (2010) carried a research on 30 Malaysian undergraduate students in
their narrative writing. The students performed the narrative writing task under no
planning, pre-task planning or on-line planning conditions. In her study, Sim (2010)
obtained similar results as in Ellis and Yuan’s (2004) and Ghavamnia, Tavakoli and
Esteki’s (2013) studies, showing that pre-task planning increased the language fluency
the most. However, mixed results were obtained from the study in terms of complexity
measures, indicating that on-line planning yielded the most number of verb form and
pre-task planning showed the highest complexity in the form of modals. The study also
pointed that planning had little effects on language accuracy. Nevertheless, Sim (2010)
did not analyse the data beyond the descriptive statistics measures and therefore, the
significance of the relationship between the variables could not be determined
adequately.
Besides Sim’s (2010) study, Seyyedi, Mohamed Ismail, Orang, and Nejad (2013)
carried out a research on 50 first-year undergraduate students in a Malaysian university.
In the research, students were assigned into two groups: planned and unplanned. Their
study revealed that planning was beneficial in improving students’ lexical complexity,
accuracy and fluency and the differences were significant except for the accuracy
measures. With the results from their study, they provide an overall picture of how
tertiary school students in Malaysia would perform when they were provided with the
opportunity to plan before performing writing tasks.
Univers
ity of
Mala
ya
49
From the studies obtained from the late 1980s until recent years, it is evident that
the study regarding the effects on planning in language production is still an ongoing
inquiry. This can be seen clearly from the inconsistency of the findings regarding the
effects of planning in the CAF of language production. Except for the measures of
fluency which majority of the research supported the positive relationship between pre-
task planning and language fluency, measures of other language constructs such as
language accuracy and language complexity and their relationship with planning still
remains as unsolved questions. The inconsistency of findings found from the previous
studies calls for the need of more empirical studies to investigate this topic. Therefore,
the present study wishes to find an answer for the question out of the inconsistency of
findings drawn from previous studies. In this section, the researcher has presented
various studies which employed different task types, genres and planning conditions.
The purpose of presenting these studies was to highlight how learners may perform
differently in different task types, genres and planning conditions and these factors
should also be considered while selecting the suitable task types, genre and planning
conditions to be employed in the present study.
2.5 Research Related to Learners’ Perceptions and Task Performance
In comparison with the studies on the effects of planning on performance,
studies which investigated learners’ perceptions of planning processes are relatively
fewer. In studies carried out by Ellis and Yuan (2004) and Li, Chen and Sun (2015), the
investigation of learners’ perceptions only served as a subsidiary aid in interpreting the
main results, i.e. learners’ performance in each planning condition. The investigation on
learners’ perceptions also helped them to see whether learners’ perceptions had
connection with their actual performance. In Ellis and Yuan’s (2004) study, they
Univers
ity of
Mala
ya
50
conducted an open-ended questionnaire survey which intended to explore the learners’
feelings on the task as well as how they approached the writing task. For the perception
part of their questionnaire survey, the findings revealed that learners did not feel
anxious whilst performing the written task. However, it was reported that the learners in
the pre-task planning and no planning groups felt some time pressure to complete the
task. In a more recent study conducted by Li, Chen and Sun (2015), the participants in
their study answered a questionnaire form which contained questions on the effects of
planning conditions of their performance. The results from Li, Chen and Sun’s (2015)
study showed that most of the participants admitted the positive effects of planning on
CAF. However, the task used in their study was an oral task. Moreover, a multiple-
choice questionnaire survey conducted by Jeon et. al. (2014) which was carried out to
investigate affective factors related to L2 writing indicated that learners’ perceptions on
language were not related to their actual proficiency level and learners’ with higher
language proficiency tended to have higher anxiety on their writing. Nevertheless, the
study conducted by Jeon et. al. (2014) had no relation with planning conditions which
are the important variables in the present study.
2.6 Identified Research Gaps
The present study aims to address the research gaps identified from the previous studies.
First of all, from the available literature, it is evident that the findings regarding the
effects of planning on CAF are still inconsistent. In the present study, the researcher
carries out the research based on the study of Ellis and Yuan (2004) due to the
adaptability of the research. The research design drawn from Ellis and Yuan’s (2004)
study also fulfils the empiricality of research which is needed to address the present
research gaps.
Univers
ity of
Mala
ya
51
Second, the available literature also shows a scarcity of research carried out
among young learners. Most of the previous research was carried out in tertiary school
contexts and hence, there might be a possibility that the effects of planning on young
learners’ writing are different than the adult writers’. The present study serves as a
necessary platform to address this research gap and at the same time, explain the
discrepancy found between the findings obtained from adult learners and young learners.
Third, the reviewed literature also displayed a lack of relevant research in
Malaysian contexts. This concern is in line with the concern of Matsuda, Canagarajah,
Harklau, Hyland, and Warschauer (2003), saying that there are still a lot of things to
discover in regards to the rhetorical, syntactical and lexical features of different genres
of writing by different learners in their distinct contexts. Furthermore, studies carried
out by Sim (2010) lacks of statistical power to prove the significance of the relationship
between planning and writing quality. Moreover, Seyyedi, Mohamed Ismail, Orang and
Nejad’s (2013) did not research on on-line planning, leaving the piece of on-line
planning remained unresearched in Malaysian contexts. Hence, this study addresses this
research gap that calls for the need to research on pre-task planning and on-line
planning in a more statistically rigourous and more holistic manner.
Other than that, the present study also attempts to fill in the research gap based
on the trend of the studies of planning. From the movement of research from the late
1980s until recent years, it is apparent that the current attention of the studies deals with
the effort of finding some consistency in contradiction of findings found from the
previous studies. Besides, there were more and more studies carried out in recent years
to investigate effects of planning on second language writing in comparison to the years
when relevant studies were just newly emerged. Hence, the present study is carried out
Univers
ity of
Mala
ya
52
due to the growing concern of researchers in investigating the effects of planning on L2
writing.
2.7 Conclusion
In this chapter, the important constructs and concepts in the present study, such as
planning and CAF have been discussed. Besides, this chapter also explained and
elaborated the theoretical framework used in the present study. Furthermore, this
chapter reviewed previous studies related to the investigated topic and from the
reviewed studies; some research gaps have been identified. With the critical discussion
of the reviewed literature and the research gaps, the needs and the importance of the
present study become clearer. In the next chapter, the methodology used to carry out the
present study will be discussed.
Univers
ity of
Mala
ya
53
CHAPTER THREE: METHODOLOGY
3.0 Introduction
In this research, the effects of planning conditions in the language performance of
Chinese primary school students were examined. The language performance of Chinese
primary school pupils was analysed in terms of complexity, accuracy and fluency
(CAF).
In any research, methodology is an important aspect which ensures the research
is carried out properly. Thus, this chapter provides a detailed description of how the
present study was carried out. The details of the study such as the participants, variables,
research design, tasks, instrumentation, data collection methods and data analysis
methods will be presented in this chapter.
3.1 The Pilot Study
Prior to the present study, a pilot study was carried out to test out the methodology and
the instruments employed in the present study. The pilot study was carried out in a
smaller scale, which involved a smaller number of participants. The pilot study aimed to:
a) obtain an earlier overview of how planning conditions affect the
language performance of Chinese primary school pupils in terms of CAF.
b) examine the appropriateness of data collection methods and instruments
to investigate the effects of planning.
Univers
ity of
Mala
ya
54
3.1.1 Participants in the Pilot Study
The pilot study was carried out with a smaller number of participants. They were 33
Chinese primary school pupils in Year 6 with the average age of 12. To the researcher’s
knowledge, all of the pupils involved in this pilot study were from Chinese background
whose first language is either Mandarin or local Hokkien dialect. English language is
the pupils’ second language. The researcher considered English language is the
participants’ second language based on the statement by Mitchell, Myles and Marsden
(2013) that ‘second languages’ are the languages learned after the acquisition of the first
language. Hence, a ‘second language’ may be a person’s third, fourth or fifth language
(Mitchell, Myles and Marsden, 2013). Furthermore, it is important to note that the
researcher would refer the participants as young learners, as they were below 18 and
fulfilled G. Ellis’ (2013) definition of young learners as children below 18 years old.
3.1.2 Procedures of the Pilot Study
The participants were equally assigned into three groups: pre-task planning (PTP), on-
line (OLP) and control group (NP) based on their proficiency level in their school
English language evaluation. The instruments used in the pilot study included a writing
task (a story consisted of 3 pictures provided with keywords) and questionnaire. The
task used in the pilot study was adapted from the UPSR trial exam question from the
Johor State Education Department (see Appendix 1).
In the pre-task planning group, the participants were provided with 7 minutes of
pre-task planning time and 10 minutes for task performance. Participants from the on-
line planning group were given unlimited time to write the written narrative without
performing pre-task planning. The participants in the control group were provided with
17 minutes of writing time without performing any pre-task planning. In the pilot study,
Univers
ity of
Mala
ya
55
pupils in all the groups used the same task sheet as appended in Appendix 1. The
measurements of CAF in the pilot study are as follows:
1. Fluency
a) Syllables produced divided by the time (per minute) used to perform the
writing task.
2. Complexity
a) Syntactic complexity: by dividing total number of clauses by T-units.
According to Hunt (1970), a T-unit is one main clause attached or
embedded with subordinate clause or nonclausal structure. In other
words, a simple or complex sentence would be considered as one T-unit,
whereas a compound sentence would be two or more T-units, depending
on the number of main clauses embedded in the sentence (Hunt, 1970).
b) Syntactic variety: total number of different grammatical verb forms used
in the task.
c) Mean segmental type-token ratio (MSTTR): ratio of total number of
different words to the total number of words in the segment (40 words).
Means scores obtained from each segment will be added and divided by
number of segments.
3. Accuracy
a) Error-free clauses: the percentage of clauses that do not contain any
syntactical, morphological and lexical choice errors.
b) Correct verb forms: the percentage of accurately used verbs which
includes the use of tense, aspect, modality and subject-verb agreement.
After performing the writing task, the participants filled in the questionnaire
forms and answered a few open-ended oral questions. The collected data was analysed
Univers
ity of
Mala
ya
56
using one-way ANOVA (analysis of variance) tests. The software used to analyse the
data was SPSS (Statistical Package for Social Science), version 16.0.
3.1.3 Results of the Pilot Study
The results of the pilot study showed that PTP had the greatest impact on young learners’
writing in terms of fluency and the difference of the mean score obtained by the PTP
group was significant in comparison with the OLP and NP groups. However, the results
showed in the complexity measures were more complicated. From the measure of
syntactic complexity, it seemed that the NP group yielded more complex sentences in
comparison with other groups and the difference was significant. In terms of syntactic
variety, the PTP group appeared to be a group which produced grammatical verb forms
with greater variation in the task but the difference in comparison with other groups was
not significant. The result obtained from the MSTTR also favoured the PTP group in
comparison with other groups and the difference was almost significant. For both the
accuracy measures, the results indicated that PTP was the best condition to produce
accurate writing. However, the scores obtained by the PTP group in both measures did
not show significant difference in comparison with other groups.
From the questionnaire survey, it was found that pupils from the planning
groups had fewer mental processes to engage with whilst performing the writing task.
Pupils from the PTP group also displayed more confidence with their performance in
writing, even though all the mean scores obtained from all items did not show any
significant difference across groups. From the questionnaire survey, the researcher
explained that the pupils felt they were given more advantage as they were given extra
time to plan the task and it boosted the pupils’ confidence on their writing. From the
reliability analysis, all the items used in the questionnaire obtained alpha values which
Univers
ity of
Mala
ya
57
were higher than 0.7 in the measure of cronbach alpha, indicating that those items were
of acceptable reliability.
3.1.4 Suggestions for the Main Study
Upon the implementation of the pilot study, the researcher discovered some weaknesses
which could be improved in the main study. First of all, there should be more
participants in the main study so that the findings found from the main study are more
generalisable. Second, the measurement of fluency can be enriched by adding in another
measure, number of dysfluencies. The measure can be accomplished by counting the
number of reformulated words and divide it by the number of produced words. Third,
the researcher found a need to include more questions in the interview session to
confirm and strengthen the results obtained from the questionnaire.
3.2 The Main Study
Drawing from the suggestions from the pilot study, the present study was carried out
with some amendments in terms of instrumentation, procedures and methodology. The
details of the main study will be discussed in the following sections.
3.2.1 The Participants
In the main study, 78 participants studying at a Chinese primary school in Klang district,
Selangor were selected. Most of the participants come from Chinese background and
their L1 is either Mandarin Chinese or the dialect of Hokkien in Chinese. However,
there are four pupils from the participants whose L1 is Malay language. The researcher
selected a Chinese vernacular school for the present study due to the linguistic challenge
Univers
ity of
Mala
ya
58
that the Chinese school pupils need to face. They need to master three languages, i.e.
English, Malay and Mandarin and in their English language writing, they cannot apply
certain rules in Mandarin and Malay (Darus & Ching, 2009). Generally, English
language is the L2 of these pupils as Mitchell, Myles and Masden (2013) claimed that a
second language is any language learned later than the acquisition of the first language.
Other than this, the participants were considered as young learners as they were at the
age of 12 in average and fulfilled G. Ellis’ (2013) definition of young learners (children
below 18 years old).
The participants in the main study were selected from Year 6 classes in the
aforementioned Chinese primary school based on the rationale that they were more
familiar with the task compared to other age group of pupils as they had been exposed
to this task format for at least 3 years and they could elicit more output for analysis. The
language proficiency of pupils who participated in the main study ranged from lower-
intermediate (50 marks to 60 marks) to high proficiency pupils (80 marks and above).
The reason for omitting pupils with basic level of proficiency was to ensure learners’
language output could be obtained adequately for analysis. Hence, the sampling method
employed in the present study is purposive sampling as the participants were selected
based on their proficiency level. It is a deliberate selection of participants due to the
qualities they possess (Etikan, Musa & Alkassin, 2016).
To determine the level of language proficiency of the pupils, the pupils were
selected based on the performance of their UPSR trial exam. All of the pupils who
participated in the present study obtained scores higher than 50 marks in their UPSR
trial exam. Other than that, the pupils were also assigned into PTP, OLP and NP groups
based on their examination marks to ensure that all the pupils in each group had the
same distribution of pupils with high, intermediate and lower-intermediate proficiency
Univers
ity of
Mala
ya
59
levels. In the present study, there were 9 pupils with high proficiency, 53 with
intermediate proficiency and 16 pupils with lower-intermediate proficiency levels. To
distribute the pupils equally into three groups, the pupils were ranked according to their
marks in the trial exam and were assigned to the PTP, OLP and NP groups in a zigzag
manner. For instance, the first to the third ranked pupils were assigned to the PTP, OLP
and NP groups and the forth to the sixth were assigned to the NP, OLP and PTP groups.
The same pattern of participants’ assignment of group was done to the 78th participant
in the present study.
3.2.2 Variables
In the main study, the independent variable was the planning conditions which pupils
attended as they performed the note expansion task. The dependent variable was the
language performance of the pupils, i.e. CAF.
The planning conditions used in the study are pre-task planning (PTP), on-line
planning (OLP) and no planning (NP). The measures of CAF used in the present study
are described as follows:
1. Fluency
a) Syllable per minute: It is a measure of fluency done by counting
number of syllables produced divided by time (minutes) used to
complete the writing task.
b) Word reformulation: It is defined as the number of reformulated words
divided it by the number of produced words. In order to obtain the
reformulation of the words correctly, the pupils were forbidden to use
Univers
ity of
Mala
ya
60
erasers or correction liquid to correct their writing. Instead, they had to
strikethrough the word, phrase or sentence that they wanted to rectify.
2. Complexity
a) Syntactic complexity: The syntactic complexity of learners’ language
can be calculated by finding the ratio of clauses to T-units. A T-unit is a
language production unit which consists of one independent clause with
its attached dependent clause(s). In the case of compound sentences, they
contain two or more T-units (Richards & Schmidt, 2002). It is a term
coined for measuring and analysing written language.
b) Syntactic variety: Syntactic variety is defined as the total number of
different grammatical verb forms used in language production. In this
research, the grammatical verb forms such as tense (e.g., simple past,
past continuous), modality (e.g. should, have to) and voice (e.g. passive
voice in the past) are taken into account in the calculation of syntactic
variety. It is considered as a measure for morphological complexity
according to Housen, Kuiken and Vedder (2012).
c) Mean segmental type-token ratio (MSTTR): For the written narratives
produced by the learners, MSTTR was obtained by finding the ratio of
total number of different words to the total number of words in the
segment (40 words). After that, means scores obtained from each
segment were added and divided by number of segments. It is a measure
used to measure lexical complexity.
3. Accuracy
a) Error-free clauses: This measure is operationalised as the percentage of
clauses that are free from syntactical, morphological and lexical choice
errors.
Univers
ity of
Mala
ya
61
b) Correct verb forms: It was counted by calculating the percentage of
accurately used verbs which includes the use of tense, aspect, modality
and subject-verb agreement.
The researcher selected these constructs for the present study due to different
rationales. The two measures for the fluency measures were chosen for the reason to
evaluate pupils’ speed in writing as well as the dysfluency occurred in their writing.
Other than this, the three measures of complexity selected in the present study
could be used to measure the grammatical and lexical aspects in pupils’ writing. In the
present study, the researcher did not investigate the interactional, propositional and
functional aspects of complexity as listed by Ellis and Barkhuizen (2005) due to the
consideration of the task used in the present study. The researcher did not need to
investigate these three aspects as the task was a narrative written task provided with
helping words, the researcher did not need to count the number of turns, idea units and
frequency of language functions as needed in measuring these three aspects.
Furthermore, the three measures selected for the present study are able to evaluate the
syntactical, morphological and lexical aspects of pupils’ written complexity, thus
evaluating pupils’ linguistic complexity more holistically besides addressing the
concern that morphological complexity is rarely researched in previous studies, as
mentioned by Bulté and Housen (2014).
Moreover, the researcher chose the two accuracy measures in order to evaluate
pupils’ written accuracy holistically and examine their accuracy specifically in the use
of verb forms. The researcher specifically looked into the use of verb forms as
according to Rozovskaya, Roth and Srikumar (2014), errors in verb forms are some of
the most general but least studied mistakes made by L2 writers.
Univers
ity of
Mala
ya
62
In order to avoid other mediating variables affecting the findings of the present
study, the pupils were not given any exposure to the constructs of CAF. However, they
were taught to use complex sentences and check their grammar in order to score higher
marks in their compositions.
3.2.3 Research Design
The present study examines the writings produced by 78 participants with the language
proficiency ranges from lower intermediate to high language proficiency levels. They
were divided into Pre-task planning (PTP), On-line planning (OLP) and control/ no
planning (NP) groups. The task conditions and the allocation of time allocated for each
group are tabulated in Table 3.1.
Table 3.1. Time allocated for each planning condition
The present study is a mixed methods research which adopts the triangulation
design. Triangulation is defined as the combination of methodologies to study the same
phenomenon (Denzin, 2009). Triangulation design is adopted for the present study as
the design is used when a researcher intends to justify or elaborate quantitative results
with qualitative data (Creswell & Clark, 2007). The design of the present study is
replicated from Ellis and Yuan’s (2004) study which consisted of quantitative and
Planning Condition n Pre-task
Planning
On-line
Planning
Time for
Writing
Pre-task Planning (PTP) 26 10 minutes 10 minutes 20 minutes
On-line Planning (OLP) 26 None Unlimited time Unlimited time
No Planning (NP) 26 None 20 minutes 20 minutes
Univers
ity of
Mala
ya
63
qualitative phases. In the quantitative phase, a quasi-experiment consisted of a post-test
was conducted where the learners were divided into three different planning conditions
before performing the writing task. There was only one post-test carried out in the
present study because this research intends to compare the differences between the
experimental groups (PTP and OLP groups) and the control group (NP group). This is
followed by the second phase of the quantitative data collection, where the researcher
administered two sets of questionnaires to the learners. In the qualitative phase, an
interview was carried out with three pupils from each group. A conceptual framework
for the research design in this study is presented in Figure 3.1.
Figure 3.1 Conceptual Framework for the Design of the Present Study
Phase 1-
Quantitative –
Quasi experiment
(Post-test design)
RQ1
Phase 2 –
Quantitative –
Questionnaire
Surveys
RQ2&RQ3
Phase 3 –
Qualitative –
Interview
RQ2&RQ3
CAF Analyses
One-way ANOVA Percentages and
frequencies
Interpretation based on the results
obtained from qualitative and
quantitative instruments.
Univers
ity of
Mala
ya
64
Furthermore, the participants in the present study were divided into three
planning condition groups: PTP, OLP and NP. The operationalisation of the writing
processes involved in each group, as described in Kellogg’s (1996) terms, is presented
in Table 3.2.
Table 3.2. Operationalisation of the writing subprocesses for all planning conditions
To create the time pressure for the pupils in the no planning group, the
researcher reduced the time used to complete the note expansion task (20 minutes) in
comparison to the time the pupils use to complete their note expansion task in the actual
examination (35 minutes). Creating the time pressure for the pupils in completing the
task was to investigate the effects of time pressure on the no planning group and OLP
group. Even though the time given by the pupils in OLP group was unlimited, the pupils
had to submit their paper as long as they finished it and the time they needed to
complete the task was recorded. For the pupils in the OLP group who did not finish the
task within the experiment time (which was about 60 minutes), their papers were
collected by the end of the experiment. Those papers were timed from the moment those
pupils started writing until the time their papers were collected.
Planning Conditions Subprocesses involved Components of Working
Memory used
Pre-task Planning (PTP) Translating
Programming
Executing
Central Executive
Phonological Loop
On-line Planning (OLP) All the subprocesses in
Kellogg’s (1996) model
Visuospatial Sketchpad
Central Executive
Phonological Loop
No Planning (NP) All the subprocesses in
Kellogg’s (1996) model
Visuospatial Sketchpad
Central Executive
Phonological Loop
Univers
ity of
Mala
ya
65
In the meantime, pupils in the pre-task planning group were given only 15
minutes to write but other than this, they were given 5 minutes to plan the task. The
total time provision allocated to the pupils in the PTP group was equal to the NP group.
The reason for this timed assignment was to ensure the time provision was fair for both
the PTP and NP groups and PTP did not take advantage over any extra provision of
time. Moreover, pupils in the PTP group were given 5 minutes to plan so that they had
ample time to write in the task. Other than that, the planning sheet of the pupils were
taken away from them when they wrote so that pupils would formulate their own
sentences without looking at the sentences they formed when they filled in the planning
sheet. The planning sheets of the pupils were also analysed to examine whether the
pupils attended to strategic planning or task rehearsal while performing pre-task
planning.
3.2.4 The Task
The task employed in the present study was a note expansion task adapted from the
actual UPSR English language paper in 2012. The task consisted of three pictures and
key words (5-6 words) were provided under each picture. The note expansion task was
about an incident describing how a boy got injured when he was flying kite with his
friends. In the note expansion task, pupils were advised to use the words provided to
describe each picture. The pupils also encouraged to write as long as they could within
the provided time. As mentioned in the previous section, pupils in the pre-task planning
group were given 5 minutes to plan and 15 minutes to write. Pupils in the on-line
planning group were allowed to spend as much time as they needed to perform the task
and the time they spent in performing the task was recorded. Besides, pupils in the no
planning group performed the task in 20 minutes without any pre-task planning. The
Univers
ity of
Mala
ya
66
task sheets used by the pupils in the PTP, OLP and NP groups are appended in
Appendix 2, Appendix 3 and Appendix 4.
3.2.5 Instruments Used in the Study/ Data Collection Methods
In the present study, there were five important instruments used in obtaining the data
needed. The first instrument used in the present study was the note expansion task sheet,
which was mentioned in the previous section. Analysing the note expansion task
performed by the pupils would provide the answers for RQ1, which were to determine
whether PTP and OLP are beneficial to young learners’ L2 writing and also to pinpoint
which planning condition is the best for L2 young writers. The second instrument used
in the present study was a questionnaire form where the pupils had to rate 13 statements
given in the questionnaire form using a 5-point likert scale. The second instrument
aimed to answer RQ 2, which was to examine how pupils attend to the narrative writing
task from the perspective of Kellogg’s (1996) model. The third instrument employed in
the present study was another questionnaire form consisted of three sections. Each
section of the questionnaire form contained statements regarding pupils’ thoughts about
their performance in each CAF construct. This instrument answered RQ 3, which was
related to the perceptions of the pupils towards their performance in terms of CAF. This
instrument also served to explain the link between planning and the perceptions of the
pupils. From that, it might provide some insights whether planning affects pupils’
psychological perceptions about their performance, even though the actual performance
might not reflect how they perceived about their performance.
Other than this, an interview was also carried out with the participants to answer
RQ 2 and RQ 3. The purpose of having the interview session was to obtain more data
about how the pupils attended to the writing task which could not have been obtained
Univers
ity of
Mala
ya
67
adequately from the first questionnaire form. Lastly, the planning sheets used by the
pupils in the PTP group were also collected to examine the pre-task planning strategies
they employed before they performed the actual narrative writing task. The result
analysed from the analysis of planning sheets could be used to describe how pupils from
the PTP group approach to the writing task, which is related to RQ 2.
The researcher designed the questionnaire and interview questions following a
few steps. Prior to the development of questions, the researcher identified the
information needed to obtain via the questionnaire and interview sessions, i.e. learners’
mental processes to approach the task and learners’ perceptions of their performance.
Then, the researcher listed out the important constructs that needed to be included in the
questionnaire and interview. By adapting the questionnaire based on Ellis and Yuan’s
(2004) study, the questions used in the questionnaire and interview were rephrased and
translated into Mandarin in order to suit the comprehension level of primary school
learners. After this, the questions used were checked and reordered for the convenience
of analysis after data collection. Based on the pilot study, it was found that there was a
need to have more questions included in the questionnaire and interview in order to
obtain richer information from the participants. The sample of the questions used in the
questionnaire and interview are attached in Appendix 5, Appendix 6 and Appendix 7.
3.2.6 Data Analysis Methods
The data collected from the pupils from each planning condition group was analysed
using the CAF analytical framework as employed by Ellis and Yuan (2004). After
analysing the texts qualitatively, basic descriptive analysis and also one-way ANOVA
(analysis of variance) were used to quantify the qualitative data. The collected data was
analysed using the SPSS (Statistical Package for Social Science) software, version 16.0.
Univers
ity of
Mala
ya
68
From the analysis using SPSS, the researcher could evaluate the effects of planning of
pupils’ CAF in writing by examining whether significant differences of pupils’
performance were manifested across groups. Besides that, the data collected from both
questionnaire forms was also analysed using the same quantitative method in order to
examine whether pupils in different planning conditions employ different cognitive
strategies and also to investigate the pupils’ perceptions on their performance after
performing the task in three different planning conditions.
Apart from that, the data collected from the interview sessions were analysed
using content analysis method. The data was analysed thematically in order to identify
the cognitive strategies the pupils used and the beliefs the pupils held when performing
the narrative writing task. The thematised contents of the interviews were later analysed
by counting the frequency of the occurrences of the thematised words in the interviews.
In order to do this, the researcher used different codes to mark learners’ response in the
interview transcripts. The codes were developed based on the mental processes listed in
Ellis and Yuan’s (2004) study and also the constructs of CAF. The list of codes used in
the interview is attached in Appendix 8.
Furthermore, the data collected from the planning sheets were categorised into
two groups, i.e. task rehearsal and strategic planning, according to the two major types
of pre-task planning proposed by Ellis (2005). The number the pupils who performed
task rehearsal and strategic planning were compared with the purpose of investigating
pupils’ preference when they perform pre-task planning.
From the collection to the analysis of data, the data collected from learners’
writings and the interview were analysed qualitatively and later represented
quantitatively. The analysis of learners’ writing was done based on the CAF framework
and the data of the interview was coded based on the list of codes listed in Appendix 8.
Univers
ity of
Mala
ya
69
To make sense of the analysis, the analysed data based on the CAF framework and
coding is statistically represented. The statistical method used to analyse learners’
writings is one-way ANOVA. Frequency count and percentage are also used to
represent interview data quantitatively. According to Ward (2007), quantifying
qualitative data enables numbers to ‘speak’ for the data. Therefore, the qualitative data
collected from the note expansion task, questionnaire forms and interview was
quantified to enhance our understanding of data.
3.2.7 Addressing the Validity and Reliability Issues of the Present Study
To address the validity and reliability issues in the present study, a few measures were
taken rigorously. First of all, the present study replicated its design and instruments
according to Ellis and Yuan’s (2004) study. According to Heffner (2015), replication is
crucial to establish validity and reliability, determine research generalisabilty and the
role of extraneous variables, apply results to real world situations and inspire new
research.
Another measure which the researcher took was to pilot the questionnaire
instruments which were used in the present study. In the field of applied linguistics,
piloting instruments helps researchers to rehearse the use of instruments, examine the
practicality of data collection methods, discover possible problems before implementing
the actual study and improve the validity and reliability of the instruments
(Rungruangthum, Todd & Aroonmanakun, 2011). The pilot study done in the present
study also served the aforementioned purposes.
Other than that, the reliability of the questionnaire items were also tested by
calculating their Cronbach’s alpha. It is a reliability measure developed by Lee
Univers
ity of
Mala
ya
70
Cronbach in 1951 to determine the internal consistency of test items (Tavakol &
Dennick, 2011). A desirable alpha value which shows acceptable reliability of an item
should be higher than .7 (Pallant, 2005). The items which had alpha values lower than .7
were rejected for the use of the present study.
Lastly, the present study also adopted triangulation data collection methods to
enhance the validity and reliability of the present study. In accordance with Holtzhausen
(2001), triangulation is a useful solution for researchers to obtain valid and reliable data.
With the triangulation of methods, the data collected would be richer and less biased. In
the present study, triangulating the data collection methods also help the researcher to
view the data in different vantage points so that more reliable results could be obtained
from the samples.
3.3 Conclusion
This chapter presented the participants, variables, research design, task, data collection
and analysis methods, as well as the ways to address validity and reliability issues.
Furthermore, this chapter also reported the pilot study conducted which contributed to
the improvement for this research. The methodological aspects of the study have been
structured in a rigorous manner in order to achieve the research objectives of this
research. The research design of the present study is similar with previous studies, such
as studies by Ellis and Yuan (2004), Sim (2010), and Piri, Barati and Ketabi (2012) in
terms of the data collection and analytical methods but the task, questionnaire forms and
interview questions were adapted for the use of young learners. In the next chapter, the
results from the data analysis will be presented in details.
Univers
ity of
Mala
ya
71
CHAPTER FOUR: RESULTS OF PLANNING EFFECTS
4.0 Introduction
In this chapter, the results obtained from the pupils’ from the data collection methods
listed in Chapter Three are presented. These results answer the research questions stated
in the first chapter.
The data of this study was collected from 78 pupils’ written texts under different
planning conditions. In addition, data was also collected from two questionnaire surveys
and interview sessions. The data of pupils’ written texts was analysed using the CAF
analytical framework as employed by Ellis and Yuan (2004). After that, the data was
analysed using statistical methods such as descriptive statistics and One-way ANOVA.
The data of both questionnaire surveys was also analysed using descriptive analysis and
One-way ANOVA whilst content analysis was used to analyse data obtained from
interviews.
4.1 Results of Independent Variables
The independent variables involved in this research are the time allocated and the
number of words and syllables produced by pupils in the three different planning
conditions. The analysis of independent variables is presented in Table 4.1. In the
analysis, the three aforementioned variables are compared in terms of means, ANOVA
and Scheffé procedures.
Univers
ity of
Mala
ya
72
Table 4.1. Analysis of independent variables
Independent
variables
Means of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-
NP
Length of time
(min.)
20 19.077 20 1.149 .322 .426 1.000 .426
Words 101.731 113.692 104.615 .908 .408 .438 .953 .621
Syllables 127.077 144.115 131.423 1.154 .321 .349 .933 .555
The results of the analysis of independent variables show that pupils in three
planning conditions spent similar amount of time to perform the task given, and
surprisingly, pupils in the OLP group took shorter time (M = 19.077) to complete the
task despite the fact that they were asked to perform the task without any time limit.
However, from the analysis of number of words, the OLP group obtained the highest
mean score (M = 113.692) in comparison to other groups although the group used the
shortest time to complete the task. Furthermore, the OLP group also produced the most
number of syllables (M = 144.115) in comparison to other groups.
In contrast with the OLP group, the PTP group used slightly longer time than
OLP group to complete the task, as they used 5 minutes to plan the task and 15 minutes
to perform it. As the provision of time for them to produce the actual writing was only
15 minutes, PTP group produced the least number of words (M= 101.731) and syllables
(M= 127.077) across groups.
Similar to the PTP group, the NP group spent 20 minutes to complete the task
without pre-task planning. The mean scores of the number of words (M = 104.615) and
syllables (M = 131.423) obtained by this group are slightly higher than the PTP group
and lower than the OLP group.
Univers
ity of
Mala
ya
73
Nevertheless, the analysis of independent variables shows that in the comparison
of mean scores across three planning conditions, the difference between the mean scores
was not significant. From that, it can be assumed that all groups were given the fair
treatment even in the provision of time, because the average time used for the OLP
group was not significantly different than the other groups. The only element that is
different across the group might be the cognitive state of the pupils while performing
the task.
4.2 Results of Dependent Variables
4.2.1 Complexity
The measurement of complexity can be examined from five distinct aspects: 1)
interactional, 2) propositional, 3) functional, 4) grammatical and 5) lexical (Ellis &
Barkhuizen, 2005). For the purpose of the present study, pupils’ written language
complexity was examined based on its grammatical and lexical aspects only, based on
studies by Crookes (1989), Mehnert (1998), Ellis and Yuan (2004) and Piri, Barati, and
Ketabi (2012).
The measurement of pupils’ language complexity consists of three parts; two of
them focus on the grammatical aspect of pupils’ writing and one of them focuses on the
lexical aspect of writing. First, the present study examines the syntactic complexity of
learners’ writing which takes account of the number of subordination of the sentences
they made. The result of pupils’ syntactic complexity is presented in Table 4.2.
Univers
ity of
Mala
ya
74
Table 4.2. Analysis of learners’ syntactic complexity
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-OLP PTP-NP OLP-NP
Syntactic
Complexity
1.096
(.114)
1.100
(.093)
1.090
(.085)
.071 .932 .990 .973 .933
From the result presented in Table 4.2, the OLP group (M = 1.1, SD = .093)
appears to be the group which yielded the highest mean score in terms of sentence
subordination. The mean score of the OLP group is closely followed by the PTP group
(M = 1.090, SD = .114), and further followed by the NP group (M = 1.090, SD = .085).
From the overall analysis, the result shows that both planning groups outperform the
non-planning group. Nevertheless, there is no significant difference shown in any of the
comparison of mean scores. The one-way ANOVA test shows that both planning
conditions have little impact on the pupils’ ability to make sentences with more
subordination. Extract 1 shows the examples of sentences yielded by the three groups of
pupils which show high and low syntactical complexity.
Extract 1
a) PTP:
Amin and his friends love to make kites. (Low syntactical complexity: 1 T-
unit and 1 clause)
b) OLP:
Ali, who is a playful boy, likes flying kites. (High syntactical complexity: 1
T-unit and 2 clauses)
c) NP:
Ali and his friends like to fly kites. (Low syntactical complexity: 1 T-unit
and 1 clause)
Univers
ity of
Mala
ya
75
Second, this study also investigates pupils’ syntactic variety under different
planning conditions. In analysing this aspect, pupils’ use of different grammatical verb
forms, including tenses, modality and voice was taken into consideration. The result of
pupils’ syntactic variety in each planning condition is presented in Table 4.3.
Table 4.3. Analysis of learners’ syntactic variety
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-NP
Syntactic
Variety
13.538
(4.254)
14.654
(3.577)
14.000
(4.391)
.488 .616 .619 .921 .847
The analysis indicates that the OLP group produced the most number of
syntactic variety (M = 14.654, SD = 3.577). The NP produced slightly lower number of
different grammatical verb forms in comparison to the OLP group (M = 14.654, SD =
3.577). Moreover, the result of the one-way ANOVA test indicates that the PTP group
demonstrated the least variation of grammatical verb forms compared to other planning
conditions (M = 13.538, SD = 4.254). Similar to the result obtained from learners’
syntactical complexity, the differences of mean scores across groups are not significant
(ρ = .616). This implies that planning conditions have little impacts on pupils’ syntactic
variety. The examples of short paragraphs from the OLP, PTP and NP groups are
presented in Extract 2.
Extract 2
a) PTP:
Last week, Ali and his friends went to a field near his house. They went to field
for flying kites. The wind was very strong. The string of the kite was broke. The
kite landed on a tree. (Number of different verb forms: 4)
Univers
ity of
Mala
ya
76
b) OLP:
Jack is a careless boy. Last Sunday was a windy day, he and his friends were
flying kites on the field happily. All of a sudden, Jack’s kite was blew by the
strong wind and landed on a tree. (Number of different verb forms: 5)
c) NP:
Last Saturday, Ali and his friends were flying their kites in the field. The wind
of that day was strong. Ali’s kite’s rope was broke and landed on the tree.
(Number of different verb forms: 4)
Third, this research also investigates the pupils’ language complexity in lexical
level. The present study employs the use measurement of the average type-token ratio in
each segment (MSTTR) which calculates the average number of different words used in
different segments (in the present study, each segment consisted of 40 words). The aim
of using MSTTR in the present study is to investigate the richness of pupils’ lexical use
under different planning conditions. The result of pupils’ lexical complexity under
different planning conditions is tabulated in Table 4.4.
Table 4.4. Analysis of learners’ lexical variety
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-OLP PTP-NP OLP-NP
Lexical Variety
(MSTTR)
.772
(.041)
.783
(.047)
.772
(.042)
.514 .600 .676 1.00 .687
The result tabulated in Table 4.4 shows that pupils in the OLP group (M = .783,
SD = .047) produced more different words in comparison to the other two counterpart
groups. Other than that, there is no difference of mean scores between the PTP group’s
(M = .772, SD = .041) and the NP group’s (M = .772, SD = .042) mean segmental type-
token ratio. However, the comparison of mean scores indicates no significant difference
across all planning condition groups in the measurement of MSTTR. The examples of
Univers
ity of
Mala
ya
77
how MSTTR was calculated for all the three groups are presented in Extract 3. Each
segment presented in Extract 3 consists of 40 words and the strikethrough words are the
words repeated in the segment.
Extract 3
a) PTP:
Last Saturday, Ali and his friends went to the big field for flying kites. On that
day was a wind day. Suddenly, Ali’s kite wire broken and the kite landed on a
big tree.
Ali climbed up to the tree (Segmental type-token ratio: 29/40 = .725)
b) OLP:
get his kite. When Brandon =climbed up the tree, the branch broke and he fell
down on the ground. He felt very painful and he cried.
=After an hour, Brandon’s friends came to the hospital to visit him. They bought
(Segmental type-token ratio: 32/40 = .8)
c) NP:
One day, Brian and his friends are flying kites in a more wind day. Suddenly,
Brian’s kite was landed on tree.
He climbed up the tree to get his kite. He stepped on a branch =and the branch
broke. Brian (Segmental type-token ratio: 29/40 = .725)
Even though there was no significant difference found in all the analysis of
pupils’ language complexity, it was still apparent that the OLP group has the highest
mean scores in the analysis of all the language complexity aspects investigated in this
study. Mixed results were found between the PTP and the NP groups as in certain
aspects, one group outperformed the other and vice versa. To conclude the analysis of
language complexity, OLP seems to be a favourable condition for the pupils to produce
more complex and variable writing. The ability to produce more complex texts is very
Univers
ity of
Mala
ya
78
important because it is closely related to pupils’ working memory capacity and
language knowledge (Hudson, 2009). Hence, the complexity of a learners’ language
may directly or indirectly reflect their language competency.
4.2.2 Accuracy
The analysis of pupils’ accuracy in narrative writing is twofold: one is on the overall
mastery of language accuracy and the other specifically focuses on the pupils’ use of
verb forms. The first measurement is the percentages of error-free clauses, which
examines clauses which are free from errors in terms of syntax, morphology and the use
of lexis. The second measurement is known as correct verb forms, which evaluates
pupils’ use of tense, aspect, modality and subject-verb agreement.
The result for the first measure, error-free clauses, is presented in Table 4.5.
Table 4.5. Percentages of error-free clauses produced by three groups of pupils
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-
NP
Error-free
Clauses
70.324
(14.837)
73.577
(15.568)
71.230
(16.940)
.293 .747 .760 .979 .867
The result shown in Table 8 indicates that OLP (M = 73.577, SD = 15.568) is
still the best condition for pupils to produce writing with better accuracy as it produced
the highest mean score in the analysis. The second highest score is from the NP group
(M = 71.230, SD = 16.940), which is slightly higher than the score obtained by the PTP
group (M = 70.324, SD = 14.837). The one-way ANOVA test also indicates that
difference of the mean percentages scored by all planning groups is too small to be
considered significant. This shows that planning conditions have little effect on pupils’
language accuracy in overall.
Univers
ity of
Mala
ya
79
For the measurement of pupils’ accuracy on the use of verb forms, the finding is
tabulated in Table 4.6. The result displayed in Table 4.6 shows an unexpected result,
with the NP group (M = 87.833, SD = 9.049) outperforming both planning groups. In
the meantime, the OLP (M = 84.258, SD = 12.538) and PTP (M = 84.111, SD = 12.189)
groups scored similar mean percentage. Taking the one-way ANOVA test result into
account, the mean scores of all the planning condition groups do not show any
significant difference.
Table 4.6. Percentages of correct verb forms produced by three groups of pupils
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-
NP
Correct verb
forms
84.111
(12.189)
84.258
(12.538)
87.383
(9.049)
.687 .506 .999 .586 .614
To conclude this part, the results obtained from error-free clauses and correct
verb forms analyses demonstrate mixed degrees of language accuracy across groups,
even though there is no significant difference found in any of the analyses. These results
infer that planning conditions may have inconsistent effects on different aspects of
language accuracy.
4.2.3 Fluency
According to Ellis and Barkhuizen (2005), there are two types of fluency measures: 1)
temporal variables, which are linked to the speed of language performance and 2)
hesitation phenomena, which are usually identified as the pauses or signs of non-fluent
language production. In the present study, both types of measures have been
investigated. The first fluency measure used in the present study is syllable per minute
and the second measure is word reformulation.
Univers
ity of
Mala
ya
80
The result obtained from the analysis of syllables per minute in pupils’ texts
shows that both planning conditions are more favourable condition to produce texts
with greater speed. From the result (see Table 4.7), in the measure of syllables per
minute, the PTP group (M = 8.472, SD = 2.691) outperformed other two groups. The
OLP group (M = 7.756, SD = 2.115) turns out to be the second most fluent group,
followed by the NP group (M = 6.571, SD = 2.252). The result of the one-way ANOVA
also indicates a significant difference in the comparison of mean scores (ρ = .017). To
be more specific, the location of significance is found in between the PTP and NP
groups (Scheffé ρ = .019).
Table 4.7. Comparison of mean scores of the measure of syllables per minute
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-NP
Syllables per
minute
8.472
(2.691)
7.756
(2.115)
6.571
(2.252)
4.282 .017 .553 .019 .203
The result from the analysis of word reformulation in pupils’ texts also
demonstrates similar finding, where both planning condition groups appear to be more
favourable conditions in avoiding dysfluencies in text production (see Table 4.8). In
both planning groups, the PTP group (M = .066, SD = .071) demonstrates lesser
occurrences of language dysfluencies compared to the OLP group (M = .097, SD
= .136). Furthermore, the NP group (M = .101, SD = .108) obtained the highest mean
score in the analysis. However, similar to most of the analyses, the comparison of mean
scores in one-way ANOVA test shows that there is no significant difference of mean
scores across groups. Hence, the finding suggests that planning conditions have little
effect in avoiding young learners’ language dysfluencies.
Univers
ity of
Mala
ya
81
Table 4.8. Comparison of mean scores of the measure of word reformulation
Dependent
variable
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-OLP PTP-
NP
OLP-NP
Word
reformulation
.066
(.071)
.097
(.136)
.101
(.108)
.799 .454 .591 .519 .993
From the findings obtained from both fluency measures, both measures suggest
that PTP promotes the language fluency of pupils. These findings are in line with the
findings obtained by Ellis and Yuan (2004), Johnson, Mercado and Acevedo (2012) and
Ghavamnia, Tavakoli and Esteki (2013).
4.3 Results of Questionnaire Surveys
The questionnaire surveys carried out in the present study consist of two parts. The first
part of the questionnaire survey aims to investigate how pupils were cognitively
engaged in the written task under different planning conditions and to further relate
these cognitive engagements to Kellogg’s (1996) Model of Process Writing. The second
part of the questionnaire survey probes pupils’ perceptions on their own performance of
the given writing task.
The result of the first part of the questionnaire survey is presented in Table 4.9.
From the questionnaire survey, it is apparent that the PTP group spent more focus on
sentence construction (M = 2.96, SD = .774), studying keywords (M = 2.92, SD = .935)
and adding details (M = 2.92, SD = 1.354). Even though the aforementioned aspects are
the aspects with the highest mean scores in the PTP group, the mean scores obtained by
the PTP group are actually lower than 3, meaning the responses are in between “neutral”
and “agree”. The general responses obtained by the PTP group in the questionnaire may
Univers
ity of
Mala
ya
82
indirectly imply that pupils in that group experienced lesser cognitive load in writing the
story and hence, less focus was given to any of the cognitive processes listed in the
questionnaire survey.
The OLP spent greater attention on sentence construction (M = 3.04, SD = .824),
grammar (M = 3.00, SD = 1.020), studying keywords (M = 2.96, SD = 1.148) and
adding details (M = 2.96, SD = 1.341). The overall mean scores obtained by the OLP
group reflect that the group paid more attention on the language structure.
On the contrary, the NP group focused more on studying keywords (M = 3.08,
SD = .891), sentence construction (M = 3.04, SD = .999) and enhancing message clarity
(M = 3.00, SD = 1.058). Based on the top three mental processes rated by the pupils in
the NP group, the processes selected showed greater diversity of activities, ranging from
the translating to editing sub-processes in Kellogg’s (1996) model.
To validate the internal reliability of the questionnaire items, a reliability
analysis was carried out. From the analysis (see Table 4.10), it is found that the
questionnaire survey is acceptably reliable (α = .760, N = 12).
Table 4.9. The result of the first part of questionnaire survey: Pupils’ focus in writing
Mental
activities
involved
M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-NP
Arranging Story 2.69
(.838)
2.73
(.778)
2.31
(.928)
1.969 .147 .987 .270 .207
Use of Words 2.73
(1.218)
2.69
(1.087)
2.42
(1.102)
.565 .571 .993 .623 .696
Sentence
Construction
2.96
(.774)
3.04
(.824)
3.04
(.999)
.068 .935 .951 .951 1.000
Grammar 2.58
(.857)
3.00
(1.020)
2.73
(1.218)
1.098 .339 .348 .868 .650
Studying
Pictures
2.77
(1.177)
2.73
(1.079)
2.77
(.863)
.012 .988 .991 1.000 .991
Studying 2.92 2.96 3.08 .167 .846 .990 .857 .917
Univers
ity of
Mala
ya
83
Keywords (.935) (1.148) (.891)
Translation 1.81
(1.357)
1.92
(1.495)
1.92
(1.294)
.060 .942 .956 .956 1.000
Considering
Sentence Types
2.27
(1.251)
2.12
(.993)
1.77
(1.243)
1.249 .293 .894 .310 .568
Clarity 2.27
(1.041)
2.42
(1.270)
3.00
(1.058)
3.032 .054 .886 .072 .190
Make the Story
Interesting
2.23
(1.210)
2.38
(1.169)
2.65
(1.198)
.838 .436 .898 .445 .719
Add Details 2.92
(1.354)
2.96
(1.341)
2.81
(1.327)
.093 .912 .995 .953 .918
Spelling and
Punctuation
2.08
(1.294)
2.38
(1.134)
2.62
(1.203)
1.292 .281 .659 .283 .791
From the findings of the first part of the questionnaire, it is clear that pupils in
different planning conditions paid different degrees of attention on different types of
mental process when they performed the writing task. The difference in the focus of
mental process might be the reason why these three groups of pupils produced different
qualities of texts.
Table 4.10. Reliability Statistics for the first part of the questionnaire survey
Cronbach’s Alpha N of items
.760 12
On the other hand, the second part of the questionnaire survey looked at pupils’
perceptions of their own performance. The pupils’ perceptions of their own performance
are compared with their actual performance in the narrative writing task. This part of the
questionnaire survey consists of three sections; each section investigates each aspect of
CAF. From the reliability statistics (see Table 4.11), the items in this part are acceptably
reliable (α = .743, N = 9).
Univers
ity of
Mala
ya
84
Table 4.11. Reliability Statistics for the second part of the questionnaire survey
Cronbach’s Alpha N of items
.743 9
Table 4.12. Pupils’ perception on their performance in terms of complexity
Descriptions M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-NP
Ability to write
long sentences
1.69
(.884)
1.88
(1.107)
1.54
(.948)
.807 .450 .781 .853 .451
Ability to use
different words
2.12
(1.033)
2.27
(.874)
1.96
(.958)
.672 .514 .846 .846 .514
Ability to use
connectors and
conjunctions
2.23
(1.142)
2.65
(.892)
2.46
(1.029)
1.108 .336 .337 .721 .796
The result of how the pupils perceive their own performance in terms of
complexity is displayed in Table 4.12. In terms of the ability to write long sentences, the
pupils in the OLP group scored the highest (M = 1.88, SD = 1.107), followed by the
PTP (M = 1.69, SD = .884) and NP (M = 1.54, SD = .948) groups. It shows that both the
planning groups have higher confidence in their written complexity when they
performed the task, in comparison with the non-planning group, even though the mean
scores scored by these three groups are low from the indicator of Likert scale, showing
that three groups have very low perception on ability to write complex sentences.
Both planning groups also have better perception on their lexical complexity
compared with the non-planning group. Following the same pattern as the previous
result, the OLP group scored the highest mean score (M = 2.27, SD = .874) followed by
the PTP group (M = 2.12, SD = 1.033) and the NP group (M = 1.96, SD = .958). In
Univers
ity of
Mala
ya
85
accordance to the Likert scale, the overall mean scores obtained by all the groups are in
the category of neutral, meaning the pupils do not have strong confidence on their
lexical complexity.
Moreover, in terms of the use of connectors and conjunctions, the OLP group (M
= 2.65, SD = .892) claim to use more connectors and conjunctions compared to the NP
group (M = 2.46, SD = 1.029) and the PTP group (M = 2.23, SD = 1.142). Even though
the use of connectors and conjunctions does not directly reflect the language complexity
of the users; some connectors and conjunctions help language users to form longer
sentences and therefore increase the chance of making more complex sentences.
However, the mean scores of three groups of pupils are slightly higher than the neutral
score, suggesting that they only have mild confidence in their use of these grammatical
items.
In parallel with the results obtained from the actual text performance, the pupils
in the OLP group have higher confidence in their language complexity. The PTP group
scored the second high mean scores in two of the three complexity descriptions in the
survey. Even though the NP group appears to be the weakest group in the questionnaire
survey, the comparison of mean scores reveals that those mean scores do not show
significant difference across group. Hence, the manipulation of planning condition may
just have slight effects on pupils’ perception in term of their confidence in language
complexity.
The OLP group also turns out to be the group with the most positive perception
on their language accuracy in comparison with other groups (see Table 4.13). In terms
of the accuracy in lexical choice, the OLP group scored the highest mean (M = 2.31, SD
= .778) followed by the NP (M = 2.08, SD = .977) and the PTP group (M = 2.04, SD
= .774). The OLP group also achieved the highest mean score in their perception of
Univers
ity of
Mala
ya
86
using correct grammar (M = 2.42, SD = .758), followed by the NP group (M = 2.23, SD
= .908) and the PTP group (M = 2.19, SD = .849). Similarly, the OLP group also scored
the highest mean score (M = 2.81, SD = 1.021) in the measure of using correct tenses.
However, the PTP group (M = 2.65, SD = .892) topped the NP group (M = 2.50, SD =
1.175) in this measure. The general feedback of the pupils from each group show that
pupils do not have strong confidence in their language accuracy, as the mean scores
obtained by each group in each description are lower than 3. There are also some
discrepancies between pupils’ perceptions and their actual performance. In the actual
performance, the OLP group scored the highest in the error-free measure and the NP
group scored the highest in the correct verb form measure. In contrast with the actual
performance, the OLP group shows stronger confidence in their performance in most of
the accuracy descriptions. The discrepancies shown between pupils’ perception and
pupils’ actual performance show that perceptions do not affect or reflect actual
performance.
Table 4.13 Pupils’ perception on their performance in terms of accuracy
Descriptions M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-NP
Using correct
words
2.04
(.774)
2.31
(.778)
2.08
(.977)
.761 .471 .525 .987 .622
Using correct
grammar
2.19
(.849)
2.42
(.758)
2.23
(.908)
.562 .572 .615 .986 .713
Using correct
tenses
2.65
(.892)
2.81
(1.021)
2.50
(1.175)
.574 .566 .867 .867 .566
For the pupils’ feedbacks on their language fluency (see Table 4.14), the NP
group obtained the highest mean score (M = 2.08, SD = .977) in the description whether
they could write smoothly in the task. The second highest mean score is owned by the
Univers
ity of
Mala
ya
87
OLP group (M = 2.04, SD = .958) followed by the PTP group (M = 1.88, SD = .766).
With the same mean scores, the PTP group (M = 1.54, SD = .811) and the OLP group
(M = 1.54, SD = .989) believe that they could write more than the NP group (M = 1.38,
SD = .941). On the other hand, the NP group (M = 2.35, SD = 1.093) scored the highest
mean score in the description whether they could write fast in the writing task in
comparison with the OLP (M = 2.31, SD = 1.050) and PTP group (M = 2.15, SD =
1.156). Despite the low mean scores obtained by all the groups in the fluency
descriptions, the survey of pupils’ perception on their language fluency yielded mixed
results and the results are contradictory to the actual performance of the pupils in their
narrative writing. Again, the results of the survey show that pupils’ perceptions do not
represent actual performance.
Table 4.14. Pupils’ perception on their performance in terms of fluency
Descriptions M (SD) of planning
conditions
ANOVA Location of significance:
Scheffé ρ
PTP OLP NP F ρ PTP-
OLP
PTP-
NP
OLP-NP
Smoothness in
writing
1.88
(.766)
2.04
(.958)
2.08
(.977)
.329 .721 .829 .747 .988
Quantity in
writing
1.54
(.811)
1.54
(.989)
1.38
(.941)
.244 .784 1.000 .833 1.000
Speed of
writing
2.15
(1.156)
2.31
(1.050)
2.35
(1.093)
.222 .801 .881 .820 .992
4.4 Results of Interview
The interview carried out on pupils were analysed in three layers. The first layer of
analysis is to investigate the mental activities of the pupils whilst performing the
narrative writing task. The second layer of analysis probes into pupils’ perception on
performance in general. The final layer of analysis examines pupils’ perception on their
Univers
ity of
Mala
ya
88
performance in terms of CAF. However, only 9 pupils were involved in the interview
due to the constraint of resource and time.
From the first layer of the analysis, it is found that all the pupils in the PTP
group agreed that they were engaged in picture comprehension and content planning.
Two-thirds of the pupils in the PTP group agreed that they involved thought about
language planning whilst only one-third of the pupils believed that they were engaged in
the use of rhetorical planning and L1 translation. Extract 4 shows how one of the pupils
in the interview reported his/her use of picture comprehension (coded as <pc>), content
planning (coded as <cp>) and rhetorical planning (coded as<rp>) strategies whilst
performing the writing task.
Extract 4
T: Oh, so you mean you are not so good at it. So what was the first thing you did
when you received the task?
P1: Look at the words. <cp>
T: Look at the words? After that?
P1: Look at the pictures<pc>.
T: Then you started writing straightaway?
P1: Uh.
T: So when you started writing, did you arrange the organisation of the story or you
did not do that because the story had been arranged readily for you?
P1: I still needed to rearrange it<rp>.
T: Oh, so you rearranged it. Okay… Okay… So did you look at the pictures given
in details? Did you study the pictures carefully?
P1: I did not study the pictures in details, just simply looked through them<pc>.
T: Just simply looked through them? So did you really count like how many kites
or people were there in the pictures?
Univers
ity of
Mala
ya
89
P1: Nope.
T: So you just looked through the pictures just to understand the flow of the story,
am I right?
P1: Mm (agrees with the statement) <rp>.
For the OLP group, all the pupils claimed that they thought about both content
and language planning, whilst two-thirds of them reported the use of picture
comprehension and similar to the PTP group, only one-third of the pupils admitted their
use of L1 translation in L2 writing. None of the pupils in this group reported the use of
rhetorical planning in performing the writing task. An example of how the transcript of
one of the pupils in the OLP group was coded is demonstrated in Extract 5. From
Extract 2, it is evident that the pupil employed the use of picture comprehension (coded
as <pc>), content planning (coded as <cp>), language planning (coded as <lp>) and the
translation from mother tongue (coded as <l1>).
Extract 5
T : Okay… what was the first thing you did when you received the task?
P3: Looked at the pictures given<pc> and then the words.<cp>
T : Ooo okay… So how did you look at those pictures? Did you study those
pictures carefully?
P3: Ummm…
T : Alright… so what was in your mind when you wrote the story? Was it the
scenes, the plot, sentence construction, words to use, grammar or you mother
tongue? What was the thing that came into your mind the most when you wrote?
P3: I was thinking of the sentences <lp> and then convert them from Chinese to
English. <l1>
T : If you compare both, which was occurring most frequently? Making
sentences or//
Univers
ity of
Mala
ya
90
P3: //I think of sentences first.<lp>
For the NP group, all the pupils also agreed that they thought of content
planning when they perform the task. Two-thirds of them reported the use of picture
comprehension, rhetorical planning and language planning. Also similar to the other
two planning groups, only one-third of the pupils reported the use of L1 translation in
their writing (see Table 4.15). Extract 6 demonstrates how a pupil from the NP group
explains his/her mental activities when s/he performed the writing task, which involved
rhetorical planning (<rp>), content planning (<cp>) and language planning (<lp>).
Extract 6
T: When you wrote the story, what was in your mind the most? Was it the
scenes of the story or how to link the story?
P: How to link the story. <rp>
T: So in between sentences, grammar, words and translation?
P: Words. <cp>
T: Okay I have a list of the things that might be in your mind when you wrote.
What was the thing that appeared to be the most prominent one?
P: Sentences. <lp>
The result of the first analysis reveals that three groups of the pupils engaged in
content planning, and only one-third of the pupils from each group agree with the
engagement of L1 translation whilst performing the task. Pupils from the PTP group
seem to put more effort in picture comprehension, content planning and language
planning whilst pupils in the OLP group put greater attention on language planning,
content planning and picture comprehension. Pupils in the NP group seem to put more
focus on content planning and equal focus on picture comprehension, rhetorical
Univers
ity of
Mala
ya
91
planning and language planning. It shows that pupils in the NP group experienced
greater cognitive load in performing the task (see Table 4.15).
Table 4.15 Mental activities involved by the pupils while performing the writing task
Planning
Groups
Mental Activities Involved in Writing
Picture
Comprehension
Rhetorical
Planning
Content
Planning
Language
Planning
Use of L1
Pre-task
Planning
3 1 3 2 1
On-line
Planning
2 0 3 3 1
No
Planning
2 2 3 2 1
The second layer of the analysis mainly addresses the pupils’ feedbacks on their
performance which were further categorised into three levels: bad, moderate and good.
The results of the analysis are presented in Table 4.16
Table 4.16. Pupils’ feedbacks on their perception of their overall performance obtained
from the interview Planning
Groups
Perception on Performance Total
Bad Moderate Good
Pre-task
Planning
1 2 0 3
On-line
Planning
0 3 0 3
No Planning 0 2 1 3
The result presented in Table 4.16 shows that two pupils in the PTP group have
moderate confidence in their overall performance whilst one of them have low
confidence in their performance. Pupils in the OLP group demonstrate higher
confidence in their quality of writing, where all of them believe that they performed
moderately in the task. Surprisingly, the NP group shows even higher confidence in
their performance, which two of the respondents suggest that they performed
Univers
ity of
Mala
ya
92
moderately in the task, while one of the respondent shows that s/he performed well in
the task.
The second layer of the analysis of the pupils’ interview transcript yields a result
that differs greatly from the questionnaire survey, which the results show that the NP
group is in disadvantage in most of the perception comparison items. The example of
how one of the pupils responded to the perception-related questions is presented in
Extract 7 and Extract 8. The code <m> indicates that they have moderate perception on
their performance.
Extract 7
T : So your mind was making sentences… How many marks will you give to
yourself for your performance, if the full marks are 10?
P3: 5. <m>
T : Okay, thank you very much.
Extract 8
T : So, how do you rate your own performance if the full marks are 10?
P6: 5. <m>
T : Okay, okay, thank you!
The final layer of analysis focuses on pupils’ perceptions of their performance
and analyses pupils’ responses according to the CAF construct. Responses related to
each construct of CAF are further divided into positive and negative responses. The
result of the analysis is presented in Table 4.17.
Univers
ity of
Mala
ya
93
Table 4.17. Pupils’ perceptions on the CAF of their writing
This layer of analysis obtained an even surprising result, which in the PTP group,
all of the respondents gave negative feedbacks on their fluency of language. In that
group, only one pupil mentioned that s/he has positive perception on language fluency.
On the brighter side, two pupils stated that they have positive perception on their
language accuracy. The pupils in the PTP group did not report anything regarding their
language complexity. Extract 9 shows how a pupil from the PTP group gave his/ her
response which signifies s/he has negative perception on language fluency (coded as<-
f>) and positive perception on language accuracy (coded as <+a>).
Extract 9
T: Oh, it’s moderately difficult. What was the most challenging thing in answering the
question do you think? Or do you think that it is not challenging at all? Or was there
any part which you were stuck there for quite some time?
P1: I didn’t know how to add in more words<-f>.
T: Didn’t know how to add in more words? Was it because I provided too little words in
the task?
P1: What I wanted to write had already been provided.
T: Huh? What you wanted to write?
P1: The words I wanted to use in the task were all being provided in the task. So I did
not what else I could use in the task<-f>.
T: What do you think were your strengths and weaknesses in completing the note
expansion task?
P1: No strengths nor weaknesses (laugh).
Planning Groups Feedbacks on
Complexity Accuracy Fluency
Positive Negative Positive Negative Positive Negative
Pre-task Planning 0 0 2 0 1 3
On-line Planning 0 1 2 1 0 3
No Planning 0 1 0 2 1 2
Univers
ity of
Mala
ya
94
T: There must be something that you are good at or bad at, such as your grammar or
your vocabulary
P1: (inaudible)
T: huh?
P1: Grammar. <+a>
For the OLP group, all the pupils mentioned that they are not confident with
their language fluency. Two respondents from the OLP group gave positive feedback
and one pupil gave negative feedback on their language accuracy. One pupil from that
group also reported negative feedback on language complexity. Extract 10 shows how a
pupil reported his/ her negative feedback on his/ her language fluency (coded as <-f>)
and at the same time, reported both positive and negative feedbacks on language
accuracy (coded as <+a> and <-a>).
Extract 10
T : What about the most difficult part of the task?
P4: The most difficult part… I don’t know.
T: What about sentence making? Or you could not express what you wanted to say?
P4: Sentence making, sentence making <-f> was a little bit difficult.
T: Okay, so what do you think are your strengths and weaknesses in performing this
task?
P4: Emmm… strengths…
T: When you were writing your task…
P4: Strengths… maybe in using past tense all these. <+a>
T: That means your grammar is good.
P4: Uh, weakness is using preposition. <-a>
The results obtained from the NP group are greatly different than the previous
planning groups as the negative feedbacks of each area of language performance are
Univers
ity of
Mala
ya
95
more than the positive feedbacks. For language complexity, one pupil reflected that s/he
has negative perception on language complexity whilst no feedback that displays pupils’
positive perception in that aspect can be found. Two pupils also reported that they have
negative perception on their language accuracy. For language fluency, two pupils gave
negative feedback and one pupil gave positive feedback on that construct. Extract 11 is
an example of how a pupil from the NP group reported his/ her negative feedbacks on
language fluency (coded as <-f>) and language complexity (coded as <-c>).
Extract 11
T: The words given? Were they easy or difficult?
P: Easy. <e>
T: Easy, so what was the most difficult part?
P: Don’t know how to add in some more words. <-f>
T: Don’t know how to add words… So what were your strengths and weaknesses in the
task just now?
P: The strength was I know how to make sentences and the weakness was I don’t know
how to add in more words.
T: Oh okay, so the sentences you made were mostly short?
P: Uh. <-c>
However, it is worth mentioning that in the interview, some pupils reported both
positive and negative feedbacks on the investigated language constructs. The
inconsistency of pupils’ response can be interpreted in two ways: 1) the pupils gave
feedbacks on different aspects of the investigated constructs, for example the use of
verb forms and prepositions under the construct of accuracy or; 2) the pupils did not
clarify their thought as they were not used to be interviewed. Extract 12 and 13 display
the occurrences of inconsistency in the interview.
Univers
ity of
Mala
ya
96
Extract 12
T: Okay, so what do you think are your strengths and weaknesses in performing this
task?
P: Emmm… strengths…
T: When you were writing your task…
P: Strengths… maybe in using past tense all these. <a>
T: That means your grammar is good.
P: Uh, weakness is using prepositions. <-a>
T: Preposition… is there anything else for your weakness?
Extract 13
T: Oh, it’s moderately difficult. What was the most challenging thing in answering the
question do you think? Or do you think that it is not challenging at all? Or was there
any part which you were stuck there for quite some time?
P1: I didn’t know how to add in more words<-f>.
T: Didn’t know how to add in more words? Was it because I provided too little words in
the task?
P1: What I wanted to write had already been provided.
T: Huh? What you wanted to write?
P1: The words I wanted to use in the task were all being provided in the task. So I did
not what else I could use in the task<-f>.
From the three-layer analysis of the interview transcripts, pupils’ descriptions of
their mental activities whilst performing the task and pupils’ perceptions on their
performance can be studied in depth. However, the results of the analysis have their
underlying reliability problems, as pupils in the age of 12 are not able to fully express
themselves and the way the interviewer interviewed the pupils might also affect how
Univers
ity of
Mala
ya
97
they responded to the questions. Nevertheless, the data obtained from the interview is
still very useful to validate the results obtained from the questionnaire survey.
4.5 Analysis of Pupils’ Planning Sheets
Other than that, the planning sheets done by the pupils in the PTP had been analysed. 26
planning sheets of the pupils were categorised according to the two major types of
planning as mentioned in Ellis (2005): 1) strategic planning and 2) task repetition.
Pupils are considered employing task repetition if their planning sheets are in full
sentences whereas pupils whose planning sheets are written short phrases, clauses or
words are considered as attending strategic planning. The analysis was done using Chi-
square test as presented in Table 4.18.
Table 4.18. Types of pre-task planning done by pupils in the PTP group
Observed N Expected N Residual
Task Repetition 14 13 1.0
Strategic Planning 12 13 -1.0
Total 26
Asymp. Sig. .695
According to the Chi-square test, the frequencies of the pupils who performed
task repetition (N=14) and strategic planning (N=12) are almost identical. Nevertheless
the number of pupils performed task repetition is slightly higher than those who
performed strategic planning. The significance level as shown in the Chi-square test
indicates that there is no significant difference between the frequencies both types of
pre-task planning.
Univers
ity of
Mala
ya
98
Nevertheless, from the study of the planning sheets, it was found that the pupils
who performed task repetition were only able to rehearse a small part of the task, as the
time given to them was only 5 minutes. The pupils who performed strategic planning,
however, were able to perform the pre-task planning more sections of the task.
Examples of how pupils performed their pre-task planning using task repetition and
strategic planning strategies are presented in Extract 14 and Extract 15.
Extract 14 (Task Repetition)
1. Last Saturday, Ali and his friends went to the field for flying kites. That day
was a wind day. Suddenly, Ali’s kite’s wires was broken and the kite landed on
a big tree.
2. After that, Ali climbed up the tree to get his kite. When just a little and can the
kite, the tree’s branch broke.
Extract 15 (Strategic Planning)
1. Last/Ali/friends/went to/field/near/house/for/wind/string/landed/tree
2.Ali/climbed/the tree/to/the kite/Suddenly/branck/broke/and/fell
down/to/ground/His friend/ sent/a/
3. came/visit/ Ali
The analysis result obtained from the Chi-square test shows that the planning
strategies employed by the pupils differ individually. There is no uniform way of how
pupils in the investigated Chinese primary school plan their essay prior to the task.
Univers
ity of
Mala
ya
99
4.6 Conclusion
From the analysis of obtained from different methods, it is apparent to see that planning
does not have significant effects on pupils’ writing, which is investigated in terms of
CAF. Besides, even though pupils in different groups are attending to similar cognitive
activities whilst performing the writing task, they have different perceptions in their
performance even though the differences are not significant. The analysis of pupils’
planning sheets also indicates that there is no specific preference of pre-task planning
strategy employed by the pupils. Detailed discussion on the results of this research is
discussed in the next Chapter Five.
Univers
ity of
Mala
ya
100
CHAPTER FIVE: DISCUSSION ON PLANNING EFFECTS
5.0 Introduction
This chapter discusses the results obtained from the present study in greater details,
which seeks to explain the reasons behind the results and links them to relevant theories.
Besides, this chapter serves an important purpose, which is to answer the research
questions based on the results obtained in the present study. Hence, this chapter justifies
the results drawn from the analyses.
5.1 The Effects of Planning on Chinese Primary School Pupils’ Written
Performance
This section answers RQ1 in the present study, which is: What type of planning has the
greater effects on Chinese primary school pupils’ written complexity, accuracy and
fluency when they perform pre-task or on-line planning?
To discuss this issue in greater depth, the effects of planning will be discussed
based on the constructs of CAF. Other than answering the research questions based on
this research, this section also compares the findings of this research with previous
studies. A discussion on the discrepancies of the comparison will provide a better
perspective on the discoveries of the present study and a clearer picture on the direction
of future studies.
Univers
ity of
Mala
ya
101
5.1.1 Planning Effects on Complexity
The measure of syntactic complexity shows that the pupils in the OLP group produced
the most complex sentences in comparison with other groups. Even though the
difference of mean scores is not significant, the result suggests that pupils who attend to
the task with no real-time pressure may produce sentences with greater complexity, as
their mind is not restricted by the time constraint. To relate it back to Kellogg’s model,
pupils in the on-line planning group had given more time to attend to both planning and
translating subprocesses. Thus, they had given the flexibility of time to retrieve the
images and information generated from their VSSP.
Apart from that, the present study also finds that pupils from the OLP group
produced writing with the most variation of verb forms and the PTP group turns out to
be the group having the lowest mean score. The inference made to explain this result is
that pupils in the OLP group were free to express what they wanted to write in the essay
as they knew they could write without any time constraint and hence, they were more
willing to take risk to use more verb forms in their writing. From that, it can also be
inferred that when the pupils are given more time to attend to both planning and
translating subprocesses, they are able to withdraw more conceptual knowledge related
to verb forms from their VSSP. The result of the study is in line with Ellis’ (2005)
proposition that unpressured on-line planning helps learners in the language formulation
by directing their selective attention to forms. Besides, Ellis (2005) also stated that PTP
facilitates learners’ conceptualisation of message but in the note expansion task given in
this research, the provision of keywords minimises the need of pupils to conceptualise
the content of their writing. Therefore, with the provision of keywords, pupils from the
PTP group tended to rely on them in order to formulate their sentences although they
had more time to plan before writing. As a result, they produced writings with low
Univers
ity of
Mala
ya
102
variation of verb forms. Nevertheless, no significant difference was found between all
the mean scores obtained in this measure due to the limitation of pupils’ language
proficiency.
Other than that, the present study also discovers that OLP condition favours the
pupils’ production of language with greater lexical variety. This result indicates that
simultaneous activation of planning and translating with no time constrain enables
pupils to generate more variation of words which are later transferred to the VSSP. In
this measure, there is no difference found between the mean scores obtained by the PTP
and NP groups. Besides, similar to the previous two measures of language complexity,
the difference of mean scores of the three condition groups is too small to be considered
significant. This can be explained by the nature of the note expansion task which some
of the important keywords were given in the task. Since the production of lexical items
is greatly determined by the pupils’ range of vocabulary, pupils’ limited vocabulary and
the provision of keywords had restricted pupils’ creativity to use different words in
delivering the content of the story.
From the three measures used in the present study to investigate the effects of
planning conditions on language complexity, the common thing in all the results is that
the OLP group scored the highest mean score in all the measures. To relate the result of
the study back to Kellogg’s (1996) model which is related to the working memory in
writing, saying that the limited capacity of central executive of the learners pushes them
to make decisions of prioritising the necessary process in writing under a time-pressured
condition. For the PTP and NP group which were bound with time limit, they had to
compromise the complexity aspect of their language in order to complete the task within
the time limit, whereas the pupils in the OLP group were able to produce more complex
language because they had lower mental pressure to perform the task and hence, their
Univers
ity of
Mala
ya
103
mind was more opened to think more creatively in writing. Thus, pupils in the OLP
group had more time and freedom to form more complex ideas and use more complex
lexical items. Even though the results show that the pupils in the OLP group spent
shorter time to complete the task in average, but in the closer look of the raw data
collected from the pupils from the OLP group, out of the 26 pupils in the OLP group, 16
pupils actually spent more than 20 minutes to complete the task and only 10 pupils
spent shorter time than 20 minutes. So in terms of the number of pupils, the number of
pupils who spent longer time to complete the task was actually more than those who
spent shorter time. To relate the result of the study back to the theoretical framework
which is related to the working memory in writing, the pupils in the OLP group were
able to produce more complex language because they had more time to undergo both
planning and translation subprocesses, which allowed them to conjure images and
conceptual knowledge and choose suitable lexical units and sentence structure to
convey the information sent to the VSSP.
Overall, the findings of the present study show that planning conditions do not
have significant impact on Chinese primary school pupils’ L2 language complexity in
writing. In contrast with the findings obtained from previous studies, such as the studies
conducted by Ellis and Yuan (2004), Piri, Barati and Ketabi (2012) and Ghavamnia,
Tavakoli and Esteki (2013), which were primarily conducted on adult learners, the
effects of planning on young L2 learners seem to be limited by major factors such as
language proficiency and also the nature of the note expansion task.
5.1.2 Planning Effects on Accuracy
The analysis of accuracy consists of two parts which investigate pupils’ use of 1) error-
free clauses and 2) correct verb forms. Even though there is no significant difference
Univers
ity of
Mala
ya
104
reported in the comparison of mean scores, the result shows that OLP group yielded the
most number of accurate clauses which conforms to Ellis and Yuan’s (2004), Seyyedi.
et. al.’s (2013) and Ghavamnia, Tavakoli and Esteki’s (2013) findings but their studies
show that PTP group produced more accurate writing than the NP group and in the
present study, the NP group performed better than the PTP group. Based on Kellogg’s
(1996) model, pupils in the OLP group had ample time to attend to all the writing
processes: 1) formulation, 2) execution, and 3) monitoring. As they were planning and
writing, they were also able to monitor and check the grammar mistakes they made and
rectify them wherever necessary.
The result of the present study also reveals that the NP group produced the most
accurate use of verb forms, followed by the OLP and PTP groups. The contradictory
results indicate that planning does not have strong effect in enhancing learners’
grammatical accuracy in L2 writing. This may be due to Chinese primary school pupils’
tendency to focus on meaning rather than the forms of language when they are planning
their writing. The explanation is in line with Sangarun’s (2005) suggestion which
highlights similar tendency of L2 learners that put their attention on the use rather than
the grammatical aspects of the language. Moreover, according to Skehan and Foster
(1997), the three constructs of CAF will compete with each other. Hence, one aspect
will be given more focus in comparison to other aspects. The claim made by Skehan
and Foster (1997) is also one of the claims postulated in Kellogg’s (1996) model: the
trade-off theory. In this study, the focus which learners paid attention to was the use of
language: fluency.
Furthermore, the PTP group scored the lowest mean scores in both accuracy
measures. This can be explained by Wendel’s (1997) argument that learners in the PTP
group were given the extra time to “off-line monitoring” their language accuracy but
Univers
ity of
Mala
ya
105
this kind of “off-line monitoring” cannot affect the on-line performance of the pupils.
As operationalised in the present study, pupils in the PTP group had relatively shorter
time to perform their task on-line (15 minutes) therefore they only managed to attend to
these subprocesses of writing: 1) translating, 2) programming and 3) executing. Hence,
pupils in the PTP group had little opportunity to check the grammatical aspects of their
writing, rendering them to compromise their language accuracy to focus on their
language fluency and complexity. The trade-off effects that took place in the PTP group
had significantly benefited the fluency of its performance especially in the measure of
syllables per minute. The PTP group also performed the best in the measure of word
reformation.
5.1.3 Planning Effects on Fluency
In line with the proposition made by previous researchers, the result of the fluency
measures shows that pupils in the PTP group produced the most fluent essays in
comparison with the other two groups in the measure of syllables per minute. The
difference of mean scores with the NP group is also significant, showing that pre-task
planning enhances Chinese primary school pupils’ written fluency. In the measure of
word reformulation, the PTP group also turns out to be the group which made the least
number of corrections in writing even though the mean score obtained by the group is
not significantly different than the other groups.
The results are similar to those obtained by Ghavamnia, Tavakoli and Esteki
(2013) and they had discussed the reasons why pre-task planning can enhance language
fluency in writing. They proposed that as pupils plan prior to the actual performance,
they engaged themselves cognitively to understand the story, organise the message
needs to be delivered and add details on the setting and characters in the task. In fact,
Univers
ity of
Mala
ya
106
the pupils in the pre-task planning stage had already attended all the subprocesses of
writing involved in Kellogg’s (1996) model if they started writing in full sentences in
the pre-task planning stage. Hence, in the actual performance of the task, pupils only
needed to access to the stored language which they had already formed in the pre-task
planning stage. From that, pre-task planning helps pupils to overcome the limitation of
working memory capacity. Since pupils in the PTP group had little cognitive load in
writing, they produced the most fluent pieces of writing in the measure of syllables per
minute and showed the least occurrences of word reformation. The results from this
research are in line with the findings obtained by Ellis and Yuan (2004), Sim (2010),
Piri, Barati and Ketabi (2012) and Ghavamnia, Tavakoli and Esteki (2013).
Furthermore, the type of pre-task planning employed by pupils also affects the
outcomes of their writing. In this study, the PTP strategies employed by the pupils were
categorised into two types: 1) task rehearsal and 2) strategic planning based on the
categorisation provided by Ellis (2005). The planning sheets which were written in full
sentences were categorised as task rehearsal sheets whilst those written in words,
chunks or short phrases were considered as strategic planning sheets. For the pupils
who performed task rehearsal, they attended to all the sub-processes of writing in
Kellogg’s (1996) model (i.e. Planning, Translating, Programming, Executing, Reading
and Editing) prior to actual writing thus they might perform better in terms of fluency in
comparison to the non-planning pupils, as the actual writing was just a repetition of the
pre-task planning. Those who attended to strategic planning are expected to produce
more fluent writing as this kind of planning helps them to establish goals, think up ideas
and organise the necessary subprocesses for writing (Ellis & Yuan, 2004). The
summary of findings related to RQ1is tabulated in Table 5.1.
Univers
ity of
Mala
ya
107
Table 5.1. Findings and interpretation related to Research Question 1
Research Questions Findings Interpretation
RQ1: What type of
planning has the greater
effects on Chinese primary
school pupils’ written
complexity, accuracy and
fluency when they perform
pre-task or on-line
planning?
1) Pupils in the PTP group
produced writing with
greater degree of fluency.
2) Pupils in the OLP group
produced writing with
greater complexity and
accuracy (error-free
clauses).
3) Pupils in the PTP group
produced writing with the
lowest degree of accuracy.
1) Pre-task planning helped
the pupils to overcome
their limitation of working
memory capacity. Pupils
who employed task
rehearsal attended all the
writing sub-processes in
Kellogg’s (1996) model
and they just had to repeat
the same writing in actual
writing; pupils who
attended strategic planning
underwent cognitive
processes such as goal
establishment, idea creation
and organisation (Ellis &
Yuan, 2004).
2) Pupils in the OLP group
had lower mental pressure
to perform the task and
they had more time and
freedom to form more
complex ideas and use
more complex lexical
items. They also had more
time to attend the
monitoring process as in
Kellogg’s (1996) model to
check their grammar
mistakes.
3) Pupils in the PTP group
only managed to attend to
subprocesses of writing
such as translating,
programming and
executing as in Kellogg’s
(1996) model, rendering
them to compromise their
language accuracy and
Univers
ity of
Mala
ya
108
5.1.4 Factors Affecting the Results of the Present Study
Based on the results obtained in the present study, it was found that most of the
comparison of mean scores in the CAF measures showed no significant difference.
Furthermore, some of the findings are similar to the previous studies despite some
unexpected discoveries. For instance, the present study found that pre-task planning aids
pupils’ written fluency and similar findings were found in Ellis and Yuan (2004), Sim
(2010), Johnson, Mercado and Acevedo (2012) and Piri, Barati and Ketabi (2012).
However, the results of this study suggest that either form of planning brings no
significant impact on L2 written accuracy. This part of findings differs from that in
Bagheridoust and Fakoor’s (2013), Salimi, Alavinia and Hosseini’s (2012) and Tavakoli
and Esteki’s (2013) studies which suggest that PTP and OLP have significant effects to
enhance the accuracy of L2 writing. Besides, the present study discovers that planning
does not have strong effect to enhance language complexity and the finding is
contradictory to the findings obtained by Ellis and Yuan (2004), Sim (2010) and
Seyyedi et. al. (2013). Other than that, all of the results obtained from the measured
items did not show significant difference of mean scores across groups, except for the
result for syllables per minute (fluency). This part of the results also greatly differed
from the results from previous studies (e.g. Skehan, 1996; Bagheridoust & Fakoor, 2013;
Ghavamnia, Tavakoli & Esteki, 2013; Yi & Ni, 2015). The contradiction of the results
found in the present study and the results obtained from the previous ones can be
accounted to many different factors. Those factors are:
focus on language fluency
and complexity.
Univers
ity of
Mala
ya
109
a) Learners’ Language Proficiency
First, the subjects of the present study are primary school pupils whilst the subjects of
the previous ones were mostly adult learners. From the perspective of their
developmental stage, the subjects of the present study might not have the range of
language proficiency which can yield results that are ‘significantly different’ than the
other learners. Thus, pupils in the primary school might not have the ability to perform
more than it is required to exhibit performance that is significantly better than others,
despite whichever planning condition they were assigned to. Therefore, learners’
language proficiency might be one of the contributing factors for the insignificant
difference of performance shown in most of the CAF measures in this study.
Different from young learners, adult learners had more exposure to the
language and the range of performance is greater than that of younger learners to show
significant difference of the results. Thus, it is easier for adult learners to obtain
significant differences in performance such as studies conducted by Ellis and Yuan
(2004), Keivan Seyyedi et. al. (2013) and Bagheridoust and Fakoor (2013). Due to the
differences of learners’ language proficiency/ developmental stages, the results obtained
in the present study should not be compared with the studies carried out on adult
learners without considering their differences.
b) Type of the Task Used in the Study
Second, the type of task used in the present study is also different from the task used in
previous studies. In previous studies such as Seyyedi, Mohamed Ismail, Orang, and
Nejad (2013) and Ghavamnia, Tavakoli, and Esteki (2013), the subjects performed their
Univers
ity of
Mala
ya
110
writing tasks with the provision of pictures as guidance. In the present study, pupils
were given not only pictures but also keywords that assist them to perform the task.
Even though the given keywords would help them to write better, the provision of them
limited the creativity of the pupils to use their own words. The provision of inflected
verbs also directed the pupils to adopt a certain type of verb tense in their writing.
Furthermore, the note expansion task used in this study was a task which pupils were
familiar with which they had been practising in their usual language lessons. The
practices and training which the pupils had in their lessons might result in the
insignificant difference of pupils’ performance as they already mastered the ways to
perform such tasks.
Moreover, a research carried out by Sangarun (2001) indicated that task
demand influenced learners’ language performance. In her research which involved
argumentative oral tasks, tasks were divided into two categories: Low and high level
cognitive and linguistic demand (LCLD and HCLD). The results of her study showed
that LCLD was more effective to direct learners’ focus towards content planning and
promoted fluency and accuracy. HCLD, however, helped to promote complexity.
Therefore, in the present study, the use of note expansion task might be the factor which
caused the insignificance of the difference of mean scores for most of the CAF
measures.
c) Time Used to Complete the Task
Third, the results of the present study are also affected by the time pupils spent to
complete the task. In previous studies, the subjects spent significantly longer time to
complete the writing task. According to Abdollahzade and Taak (2014), learrners need
to have at least 10 minutes of planning time in order to yield significant effects on
Univers
ity of
Mala
ya
111
written performance. Referring to the planning time allocated to the learners in the
present study, the PTP group only spent 5 minutes to plan whereas the OLP group did
not spend longer time to write the essays as expected. The insufficient provision of time
given to both planning groups might be the factor that made the results different from
those obtained from Bagheridoust and Fakoor’s (2013), Salimi, Alavinia and Hosseini’s
(2012) and Ghavamnia, Tavakoli and Esteki’s (2013). However, despite using shorter
duration to complete the task, the OLP group yielded the most number of words and
syllables in the study. The result obtained suggests that real time pressure might affect
the pupils’ production of writing.
d) Pupils’ Mental Activities
Other than that, the mental activities underwent by the pupils while performing the task
was one of the key factors that affected their written performance, assuming all the
groups had equal pupil distribution of different proficiency levels. Based on the results
obtained from the questionnaire survey, it was shown that pupils in the planning groups
had low average mean scores for all the items related to the possible mental activities
involved in ESL writing. Even though the NP group had higher mean scores in
comparison with the other two groups, which indicated that pupils in the group had
more mental activities to attend, the difference of the mean scores across these three
groups was actually not significant. This might prove that pupils in all the groups might
not have much difference in terms of the number of mental activities engaged despite
the results shown from the questionnaire survey. The details of the mental activities
engaged in each group will be explained in 5.2.
Univers
ity of
Mala
ya
112
e) The Planning Strategy Employed by the Pupils in PTP Group
According to the study carried out Sangarun (2001), three types of pre-task planning
were employed to examine their effects on language performance: 1) content focused
(CFPP) 2) language focused (LFPP) and 3) content and language focused (CLFPP) pre-
task planning conditions. Among all PTP conditions, content focused (CFPP) and
content and language focused (CLFPP) planning conditions are able to promote
language complexity and all types of pre-task planning conditions promote accuracy
and fluency.
Nevertheless, the results obtained in the present study contradicted with the
results obtained from Sangarun (2001). The PTP group did not outperform the NP
group significantly in all of the measures except for the measure of syllabus per minute
(fluency). Hence, a deeper investigation on how pupils in the PTP group performed
their pre-task planning has to be done.
Even though in the present study, pre-task planning conditions were not being
oparationalised in Sangarun’s (2001) terms, the study provides some insights about how
the focus of pre-task planning affects the performance of learners. In the present
research, the planning sheets obtained from the pupils were analysed into two main
types: 1) strategic planning and 2) task rehearsal. Rehearsal is the repetition of task prior
to the actual writing (Hsu, 2012) whilst strategic planning is the preparation done for a
language task by considering the encoding and expression of the actual task materials
(Ellis, 2005). Based on the definitions given by Ellis (2005), strategic planning can be
grouped as content focused and task rehearsal as content and language focused pre-task
planning. In this study, 14 pupils employed task rehearsal to plan and 12 pupils
employed strategic planning.
Univers
ity of
Mala
ya
113
From the examination of the pupils’ planning sheets, the researcher also found
that pupils did not fully perform their task rehearsal, as the time given for pre-task
planning was only 5 minutes. From all the pre-task planning sheets collected, the
researcher found that all the pupils who performed task rehearsal only managed to
describe the first picture in the task, whereby three pictures were provided in the task.
This may be a factor which caused the results obtained from the PTP group different
from the expected results.
Furthermore, one of the possible reasons which caused the insignificance of
performance was pupils’ lack of training in pre-task planning. In researcher’s
speculation, pupils in the PTP group might perform better if they were given ample
training in performing pre-task training.
5.2 The ESL Writing Processes of Chinese Primary School Pupils
In relation to RQ2, which is about how Chinese primary school pupils attend to the
narrative writing task under different planning conditions, this section discusses how the
results obtained in the present study can be explained by the hypotheses proposed by
Kellogg’s (1996) Model of Writing Processes. Unlike most of the studies on planning
effects on CAF which primarily focus on ESL adult learners’ writing process, this
discussion focuses on the writing processes of young learners of ESL, or more
specifically Chinese Primary School ESL pupils.
5.2.1 Trade-off Effects
Other than the aforementioned issues, one of the linguistic phenomena that is closely
related to the present study and in some degree affects the results is the trade-off effects
Univers
ity of
Mala
ya
114
that occurred in pupils’ mind while attending to the task. It has been argued that learners
are not able to focus on all three areas of CAF, especially when the task is challenging
to the learners (Vercellotti, 2012). In this case, the results of the present study indicate
that pupils do have a certain degree of trade-off under different task conditions.
Within the three constructs of CAF, previous studies suggest that there is a
trade-off between language fluency and accuracy (Ahmadian & Tavakoli, 2011; Yuan
& Ellis 2003; Michel, Kuiken & Vedder, 2007). The similar finding can be applied to
the present study which indicates that in the PTP condition, pupils are able to produce a
more fluent writing but not a more accurate one. Besides, the OLP condition seems to
be able to produce more complex essays even though the effects are not as significant as
the effects made by the PTP group on written fluency. Most of the pupils in the OLP
group needed a longer duration of time to perform the task, indicating that pupils in the
OLP group compromised their language fluency to enhance their language complexity.
However, both conditions do not seem to have strong effects on language accuracy.
This part of finding, especially the find obtained from the PTP group, contradicts with
Skehan and Foster’s (1996) hypothesis, suggesting that trade-off effect exists between
complexity and accuracy. Nevertheless, Ortega (1999) suggests that Skehan and
Foster’s hypothesis should take learners’ proficiency level into consideration. Ortega’s
(1999) statement explains that the trade-off effect between complexity and accuracy
will not be evident if the learners are of lower proficiency. Clearly, Ortega’s (1999)
statement answers the reason of the different findings obtained in the present study.
5.2.2 Pupils’ Foci in Writing under Different Planning Conditions
To identify the mental activities that the pupils focused on in their writing task, the
mental activities with the mean score higher than 3.00 in the questionnaire survey are
Univers
ity of
Mala
ya
115
considered as the pupils’ foci in writing. For the PTP group, none of the mental
activities obtained mean scores higher than 3.00, except for sentence construction (M =
2.96). For the OLP group, there are two mental activities that scored 3.00 and above,
which are sentence construction (M = 3.04) and grammar (M = 3.00). To classify it
using Kellogg’s (1996) model, the prominent mental activities carried out by both the
PTP and OLP group can be classified under the sub-process of translating, which is the
sub-process where the writers select the lexical units and syntactic frames used in the
texts they produce. For example in the writing task employed in this study, the pupils
have to choose the appropriate words to link the keywords given in the task to form
sentences. Moreover, they need to arrange the words according to the correct syntactical
order, for instance positioning subject, verb and object in a sentence in a grammatically
acceptable sequence.
For the non-planning group (i.e. NP group), more mental activities have been
identified, suggesting that pupils in the NP group were more cognitively loaded when
they performed the writing task. The mental activities that the pupils focused on include
studying keywords (M = 3.08), sentence construction (M = 3.04) and clarity (M = 3.00).
To explain these activities in Kellogg’s (1996) model terms, these activities involve the
sub-processes of planning (the sub-process of establishing goals, creating and
organising ideas) and translating.
The results obtained from the interview are somewhat similar to the
questionnaire results, indicating that the pupils in the NP group had relatively more
mental activities to attend whilst performing the task. From the results of the interview,
most of the pupils in the planning groups only need to attend to three mental activities:
picture comprehension, content planning and language planning. Most of the pupils in
the NP needed to engage themselves in picture comprehension, rhetorical planning,
Univers
ity of
Mala
ya
116
content planning and language planning. The findings obtained of the interview are
similar to Ellis and Yuan’s (2004) findings which suggest that learners in the NP group
had greater cognitive load whilst performing the task.
The results obtained from the questionnaire survey and interview suggest that
pupils in the planning groups have fewer mental activities to focus on and those mental
activities are classified as the translating sub-process in terms of Kellogg’s (1996)
model. Whereas for the non-planning group, pupils focused on more mental activities
and they involved all both planning and translating subprocesses with the formulation
system in Kellogg’s (1996) model. The findings obtained in this study hence agree with
the control mechanism claim suggested in Kellogg’s (1996) model, which is related to
the cognitive load of learners in writing. In this case, pupils in the PTP group allocated
their mental resources solely on the sub-process of translating, as they had rehearsed the
task prior to the actual writing, whilst pupils in the OLP group needed to focus on more
mental activities but their cognitive load was reduced due to extra time given. Hence,
pupils in the OLP group focused not only on sentence construction but on grammar
checking as well. The pupils in NP group have greater cognitive load to perform their
writing task as they have more mental activities to focus in the limited time provided.
5.2.3 Pre-task Planning Strategies Employed by Pre-task Planners
From the 26 participants in the PTP group, 14 pupils attended to task rehearsal whilst 12
pupils performed strategic planning before they performed the actual task. The result of
the study reveals that pupils’ preference of both types of pre-task planning is almost
equal. Instead, it is up to their personal decision to choose which planning strategy to be
employed in their writing. Since the numbers of pupils who performed task rehearsal
and strategic planning are almost equal and the PTP group outperformed the non-
Univers
ity of
Mala
ya
117
planning group in terms of fluency, it could be implied that both types of planning do
have some effects on pupils’ written fluency. This part of result is in line with the result
obtained by Sangarun (2001) which claims that all types of pre-task planning, regardless
of their focus, have positive impact on language fluency.
Based on the analysis of the planning sheets, the researcher found that pupils in
the Chinese primary school do not master the proper skills to plan a task before writing
it (as shown in Extracts 14 & 15, Chapter 4). This might be one of the major factors
which affect the pre-task planning strategy employed by the pupils in approaching the
task given. The pupils’ immature pre-task planning might be due to their lack of such
exposure in their daily instructions. The summary of findings related to RQ2 is
presented in Table 5.2.
Table 5.2. Findings and interpretation related to Research Question 2
Research
Question
Findings Interpretation
RQ2: How do
Chinese primary
school pupils
attend to the
narrative writing
task under different
planning
conditions?
1) Pupils in the PTP group
compromised their language
accuracy and focused on
their language fluency.
2) Most pupils in the OLP
group compromised their
language fluency to attend to
the accuracy aspect of their
writing.
3) Pupils in the PTP group
focused more on sentence
construction.
4) Pupils in the OLP group
paid more attention on
sentence construction and
1) The finding indicated the trade-off
between accuracy and fluency when
pupils were to perform the task under
time-pressured condition.
2) Pupils in the OLP group spent
their time to elaborate their writing
and to check their grammar. Hence,
trade-off effect was shown where
pupils compromised their language
fluency to improve their complexity
and accuracy.
3) Pupils in the PTP group had less
cognitive load in their writing as they
had attended the task in their pre-task
planning.
4) Pupils in OLP group had to attend
all the mental processes in Kellogg’s
(1996) model but their cognitive load
Univers
ity of
Mala
ya
118
5.3 Chinese Primary School Pupils’ Perceptions on their Different Planning
Conditions
This section of discussion answers RQ3, which is: What are the Chinese primary
school pupils’ perceptions on their performance in different planning conditions?
5.3.1 The Differences between Pupils’ Perceptions and Actual Performance
Other than looking for the connection between planning conditions and written
performance, the present study also investigates the effects of planning on pupils’
grammar.
5) Pupils in the NP group
attended to more mental
activities compared with
both planning groups.
6) Pupils in both planning
conditions indicated the use
of picture comprehension,
content planning and
language planning when
they performed the task.
7) Pupils in NP group
indicated the use of picture
comprehension, rhetorical
planning, content planning
and language planning when
they performed the task.
was reduced due to the extra time
given to them.
5) Pupils in the NP group had to
attend to all the mental processes in
Kellogg’s (1996) model under time-
pressured condition, hence their
cognitive load was heavier.
6) Pupils in both planning conditions
carried out planning process which
involved strategies as stated.
However, their planning process was
not as complicated as the process
carried out by the NP group.
7) Pupils in the NP group planned in
a more sophisticated way which
involved more strategies than the
planning groups. This finding reveals
that pupils in the NP group had
heavier cognitive load to process in
writing.
8) 14 of the pupils in the
PTP group employed task
rehearsal whilst 12 of them
used strategic planning to
plan their task.
8) Pupils in the Chinese primary
school do not have special preference
in their pre-task planning strategy.
The finding also shows that pupils do
not master the proper technique of
pre-task planning.
Univers
ity of
Mala
ya
119
perception on their performance. The present study also intends to link pupils’
perception with their actual performance. From the results of the questionnaire, pupils
in the OLP group obtained the highest mean scores for all the items in terms of
language accuracy and complexity, this part of finding reflects the actual performance
of pupils in their writing. On the other hand, pupils from the NP group reported the
highest mean scores for most of the aspects of language fluency, showing that they were
more confident in language fluency in comparison with other groups. This part of
finding differs from pupils’ actual performance as in the analysis of pupils’ writing,
pupils from the PTP group obtained the highest mean scores for the measures of fluency.
The possible reason of why the NP group obtained the highest mean score might be
their misperception of fluency in which they misconstrued “time pressure” as fluency.
The results obtained from the interview, however, are different than the results
from the questionnaire survey. The results from the interview show that pupils from all
the groups have negative perceptions in their language fluency and complexity. From
that, it can be assumed that pupils in the investigated Chinese primary school had low
confidence in their ESL writing in terms of fluency and complexity. Pupils’ low
confidence in their writing, if it is not being given enough concern, might result in
pupils’ disinterest in language learning in a long run. Thus, it is important for the
researcher to identify the effects of planning conditions on pupils’ perception. From the
findings of the interview, it is shown that both types of planning had no effect to evoke
pupils’ positive perception of their writing in terms of fluency and complexity.
Nevertheless, the pupils from planning groups are more confident with their
language accuracy in comparison with the non-planning counterpart. This part of
finding, however, greatly differs from pupils’ actual performance as reflected in the
analysis of writing. In actual performance, pupils in the OLP group scored the lowest
Univers
ity of
Mala
ya
120
mean scores in all the accuracy measures. This finding reveals that pupils’ perception is
not a strong predictor of their actual performance. However, this finding indicates that
both planning condition had positive effects on pupils, at least in perception level.
The results obtained from the questionnaire survey and interview show that
pupils’ perceptions on their language complexity and accuracy do not reflect their actual
performance. The results of the present study conform with the results of Jeon et.al.’s
(2014) study which states that learners’ perception on their writing is not related to their
actual performance in writing in terms of complexity and accuracy. However, their
research did not investigate the role of perception on language fluency and the
participants in their research were adult learners. The summary of findings which
answer RQ3 is shown in Table 5.3.
Univers
ity of
Mala
ya
121
Table 5.3. Findings and interpretation related to Research Question 3
5.4 Conclusion
This chapter discusses the results obtained in this research and it is found the results of
the present study contradict with the results of previous studies, except the part that pre-
task planning promotes pupils’ written fluency. Despite the fact that planning conditions
had not significant effects on most of the CAF measures in the present study, the study
reveals the trade-off effect between form (complexity and accuracy) and fluency. The
results of the present study obtained from questionnaire and interview also suggest that
Research
Question
Findings Interpretation
RQ3: What are the
Chinese primary
school pupils’
perceptions on
their performance
in different
planning
conditions?
1) The questionnaire results
report that pupils from the
OLP group scored the
highest for the items related
to complexity and accuracy.
2) In the questionnaire
survey, pupils in the NP
group reported the highest
mean scores for most of the
aspects in language fluency.
3) The interview results
show that pupils from all
groups have negative
perceptions in their
language fluency and
complexity.
4) From the interview, it
discovers that pupils in the
planning groups are more
confident in their language
accuracy.
1) Pupils’ perception reflected their
actual performance in terms of
complexity and accuracy.
2) In fact, pupils in the PTP group
scored the highest in terms of
fluency. Pupils in the NP group
perceived “time pressure” as
fluency.
3) This finding reflects pupils’ low
confidence in second language
writing. The finding also indicates
that both types of planning had no
effect to evoke pupils’ positive
perception of their writing in terms
of fluency and complexity.
4) Pupils in the PTP group scored
the lowest in all the fluency
measures in the actual writing. The
finding shows that pupils’ perception
does not fully reflect their actual
performance. Nevertheless, the
finding shows that both types of
planning are able to boost pupils’
confidence in their writing in terms
of accuracy.
Univers
ity of
Mala
ya
122
planning conditions reduced pupils’ cognitive load in performing the writing task.
However, the present chapter also discusses the factors associated with the insignificant
difference of mean scores in most of the measures of CAF which include pupils’
language proficiency, type of task used, time spent to accomplish the task, pupils’
mental activities and the PTP strategy used by the pupils in the PTP group. The present
study also discovers that despite implementing the task under different planning
conditions, pupils in all the group had low confidence with their performance in general.
Moreover, the findings also indicate that pupils’ perception only partially reflected their
actual performance. Since the present study deals with young learners, the present study
calls for more research on young learners to investigate their L2 writing behaviours.
Univers
ity of
Mala
ya
123
CHAPTER SIX: CONCLUSION
6.0 Introduction
This chapter presents the overall summary for the whole study and discusses the
implications of the findings in the contexts of English language teaching and
assessments. This chapter also provides a few suggestions for future research based on
the issues raised in the present study.
6.1 Summary of the Present Study
According to the findings obtained from the present research, it is apparent that pupils
in Chinese primary school can produce more fluent writing if they are provided with the
chance to plan before performing the writing task. On the other hand, if the pupils are
given the chance to write without time pressure, they are more likely to produce writing
with greater complexity and accuracy. However, from the statistical analysis, the effects
yielded by both planning conditions are not significant in comparison to their non-
planning counterpart, except for the effect of pre-task planning on the fluency of writing
measured by syllables per minute.
Based on the questionnaire survey carried out amongst the pupils, the researcher
discovers that pupils in both planning conditions have considerably fewer mental
activities to focus on in comparison to those who performed under the non-planning
condition. This can be related to the theoretical framework employed in the present
study, Kellogg’s (1996) model. Kellogg’s (1996) model proposes that writers in
planning conditions have lesser cognitive load in writing and therefore, they can
perform better than the non-planners.
Univers
ity of
Mala
ya
124
Other than this, the present study discovers that there was some degree of trade-
off effects taking place among the pupils in both planning conditions as they performed
the writing task. In the pre-task planning condition, pupils tend to focus more on
language fluency and in the on-line planning condition; pupils tend to yield more
complex language rather than the pupils who did not plan at all. The results of the
present study slightly contradict with the studies carried out by Ahmadian and Tavakoli,
(2011), Yuan and Ellis (2003) and Michel, Kuiken and Vedder (2007) that suggest
trade-off occurs between fluency and accuracy of language production.
To further investigate how pre-task planning affects pupils’ written production,
the present study also examined how pupils performed their pre-task planning. The
result of this study reveal that the number of pupils performed strategic planning and
task rehearsal is almost equal. This shows that there is a possibility for both types of
PTP strategy to have some impact on pupils’ quality of writing, especially in the aspect
of language fluency as discovered in the present research. Moreover, the analysis of
PTP sheets also indicates that Chinese primary school pupils do not master the correct
skills to perform pre-task planning.
Moreover, the results of the present study also discovered that pupils in different
planning condition also have different perceptions on their writing performance, and
their perceptions can somehow predict their actual performance in terms of language
accuracy and complexity but not in terms of language fluency.
6.2 Theoretical Implication of the Present Study
The present study provides better theoretical understanding on how planning affects
young ESL learners’ writing. In the present study, it is found that pre-task planning
Univers
ity of
Mala
ya
125
enables pupils to write more fluently and on-line planning tends to enhance pupils’
language complexity and accuracy. The findings obtained in the present study give SLA
researchers an insight that planning conditions may yield different language
performance depending on the learners’ developmental stage, as the present study is
carried out on young learners instead of adult learners in previous studies.
Furthermore, the present study investigates how young learners approach the
given writing task. The findings of the present study show that there is a trade-off in
between fluency, accuracy and complexity depending on pupils’ planning conditions.
Pupils who performed pre-task planning tended to compromise their language accuracy
to favour the fluency aspect whereas those who performed OLP focused more on
language accuracy and complexity rather than fluency. The present study also shows
that pupils who are under no planning condition experienced heavier cognitive load as
they needed to attend to all mental processes as indicated in Kellogg’s (1996) model.
These findings are similar with those obtained in Ellis and Yuan’s (2004) study.
However, the present study discovered that more than half of the participants in the PTP
group performed task rehearsal instead of strategic planning in their pre-task learning
stage. The participants’ preference of pre-task planning might be different from that of
adult learners. Nevertheless, the findings of the present study call for more research to
investigate young learners’ mental processes in performing second language writing
tasks as well as how these mental processes ultimately affect their performance.
6.3 Pedagogical Implications of the Present Study
As indicated in the findings obtained from the present study, it is found that pupils who
performed under OLP condition were better able to produce writings that are more
complex and accurate. An implication is that ESL teachers in primary schools should
Univers
ity of
Mala
ya
126
coach pupils the correct ways of monitoring and editing their own writing in order to
maximise their performance.
Furthermore, ESL teachers should also teach the pupils the correct ways of
planning in order to enhance pupils’ fluency in writing, as the findings from the present
study show that pupils in the PTP group were able to write more fluent essays in
comparison with other planning groups. In this case, teachers should teach them
strategies of outlining their writing prior to their actual writing, instead of performing
task rehearsal as it would take time and pupils would not have ample time to rehearse
the whole task in actual evaluation settings.
Other than that, teachers should also design tasks or test items that encourage
primary school pupils to employ both PTP and OLP as both planning conditions
complement each other in enhancing the performance in terms of CAF. Employing both
pre-task and online planning would help pupils to improve their performance as pre-task
planning helps pupils to focus on meaning whilst on-line planning encourages pupils to
focus on form.
Moreover, teachers in primary schools should also be mindful of the
psychological effects of planning in pupils’ perception in writing. As the pupils are
given the opportunity to plan, they will have more positive perceptions on the outcomes
of their writing. By improving pupils’ perceptions in writing, it is hoped that pupils are
motivated to write and ultimately improve their proficiency in writing.
6.4 Suggestions for Future Studies
The present study shed light on how younger learners, especially those in the Chinese
primary schools context, perform writing tasks under different planning conditions.
Univers
ity of
Mala
ya
127
However, since very few studies have been carried out in the context of young learners
of ESL, the present study calls for more research to be carried out on these learners.
For future studies, researchers can carry out similar experiments on larger
population to ensure the findings can be generalised to larger population. Furthermore,
pupils from different schools can be selected for future study to increase the
applicability of the findings to other primary schools in Malaysia.
In order to increase the generalisability of the findings obtained in the present
study, similar studies can also be carried out to young learners of different proficiency
levels. In the present study, the researcher controlled the level of proficiency of pupils in
order to ensure pupils could elicit ample language for analysis. In future studies,
participants can be categorised as low, intermediate and high proficient learners.
Moreover, researchers in future studies can address the issue of time on pupils’
implementation of task, as the pupils in the OLP group in the present study did not use
more time to implement the task as expected. It may also be one of the reasons that the
results obtained by that group were not significantly different than the non-planning
group.
Other than that, future studies can also investigate primary school pupils’ oral
production using the same analytical framework and research design. Similar research
have been carried out in other learners’ contexts using theoretical frameworks such as
Levelt’s (1989) model of speech production but primary school context is still a very
unexplored learning context.
Furthermore, future studies can also include another experimental group that
performs under PTP and OLP conditions to investigate whether this combination can
further enhance pupils’ written performance. Piri, Barati and Ketabi (2012) have
Univers
ity of
Mala
ya
128
included this group in their study but their research did not include a control group,
which is the NP group. As a suggestion, future studies can include a control group in
order to make sense of the effects of three planning conditions (PTP, OLP and
PTP+OLP) on written performance.
In the present study, the research also discovered that pupils engaged in both
pre-task planning types, i.e. task rehearsal and strategic planning. Hence, there is also a
possibility for researchers to compare the effectiveness of both types of pre-task
planning in future studies. Investigating this aspect can provide an insight on the type of
pre-task planning that is more effective in enhancing pupils’ written performance.
Moreover, the present study also calls for more investigation on the longitudinal
effects of planning on young learners’ performance, as the present study is a cross-
sectional study. Carrying longitudinal studies can provide better insights for researchers
regarding the role of planning in developing young learners’ writing proficiency.
In future studies, researchers should also consider training the participants in the
PTP group to perform pre-task planning prior to the task. The rationale of training the
participants to perform pre-task training is to ensure that the participants are able to
perform the “correct” way of pre-task planning, so that future researchers can ascertain
whether pre-task planning can yield results with significant difference in comparison
with other planning conditions.
Last but not least, researchers can also investigate other forms and genres of
written tasks in future studies as the present study focused only on the note expansion
task and the genre of narrative writing. Nevertheless, researchers should consider
whether the pupils are familiar with the chosen written task as researchers need to
obtain pupils’ naturalistic output that can represent pupils’ real performance.
Univers
ity of
Mala
ya
129
6.5 Conclusion
All in all, the present study obtains results that are partially different than those obtained
in previous studies. From that, the present study provides an insight on planning
conditions affect young learners’ written performance. The findings are valuable as
previous studies rarely researched on young learners and the present study addresses
this research gap. Nevertheless, due to the limitations mentioned in this study, there are
still a lot of possible and relevant angles to approach the writing of Chinese primary
school pupils.
Univers
ity of
Mala
ya
130
REFERENCES
Abdollahzade, Esmaeel, & Taak, Morteza. (2014). Planning time and writing quality in
expository writing. Procedia - Social and Behavioral Sciences, 98(0), 29-35. doi:
http://dx.doi.org/10.1016/j.sbspro.2014.03.385
Abdollahzadeh, S., & Kashani, A. F. (2011). The effect of task complexity on EFL
learners' narrative writing task performance. Journal of English Language
Teaching and Learning, 8(1), 1-26.
Alamargot, D., & Chanquoy, L. (2001). Through the models of writing: With
commentaries by Ronald T. Kellogg & John R. Hayes: Springer Netherlands.
Ali, Nor Liza, Hamid, M. Obaidul, & Moni, Karen. (2011). English in primary
education in Malaysia: Policies, outcomes and stakeholders’ lived experiences.
Current Issues in Language Planning, 12(2), 147-166. doi:
10.1080/14664208.2011.584371
Baddeley, A.D. (1992). Working memory. Science, 255(5044), 556-559. doi:
10.2307/2876819
Baddeley, A.D. (2004). The psychology of memory. In A. D. Baddeley, M. D.
Kopelman & B. A. Wilson (Eds.), The Essential Handbook of Memory
Disorders for Clinicians: John Wiley & Sons, Ltd.
Baddeley, A.D., & Logie, R.H. (1999). Working memory: The multiple component
model. In A. Miyake & P. Shah (Eds.), Models of Working Memory:
Univers
ity of
Mala
ya
131
Mechanisms of Active Maintenance and Executive Control. Cambridge
University Press.
Bagheridoust, E., & Fakoor, K. A. (2013). The effects of planning on accuracy in
argument/compare and contrast writing of Iranian EFL learners. International
Research Journal of Applied and Basic Sciences, 4(12), 3764-3773.
Becker, A. (2006). A review of writing model research based on cognitive processes. In
A. Horning & A. Becker (Eds.), Revision: History, theory, and practice: Parlor
Press LLC.
Bereiter, C., & Scardamalia, M. (1987). The psychology of written composition: L.
Erlbaum Associates.
Berninger, V., Whitaker, D., Feng, Y., Swanson, H. L., & Abbott, R. D. (1996).
Assessment of planning, translating, and revising in junior high writers. Journal
of School Psychology, 34(1), 23-52. doi: http://dx.doi.org/10.1016/0022-
4405(95)00024-0
Bley-Vroman, R. (1983). The comparative fallacy in interlanguage studies: The case of
systematicity. Language Learning, 33(1), 1-17. doi: 10.1111/j.1467-
1770.1983.tb00983.x
Univers
ity of
Mala
ya
132
Bulté, B., & Housen, A. (2014). Conceptualizing and measuring short-term changes in
L2 writing complexity. Journal of Second Language Writing, 26, 42-65. doi:
http://dx.doi.org/10.1016/j.jslw.2014.09.005
Burke, A. (2012). Empowering children's voice through the narrative of drawings. The
Morning Watch: Educational and Social Analysis, 40(17), 1-14.
Cho, Y. (2003). Assessing writing: Are we bound by only one method? Assessing
Writing, 8(3), 165-191. doi: http://dx.doi.org/10.1016/S1075-2935(03)00018-7
Connor, U. (1996). Contrastive rhetoric: Cross-cultural aspects of second language
writing: Cambridge University Press.
Cortazzi, M. (2014). Narrative analysis: Taylor & Francis.
Creswell, J. W. (2012). Educational research: Planning, conducting, and evaluating
quantitative and qualitative research (Fourth ed.). United States: Pearson.
Creswell, J.W., & Clark, V.L.P. (2007). Designing and conducting mixed methods
research: SAGE Publications.
Crookes, G. (1989). Planning and interlanguage variation. Studies in Second Language
Acquisition, 11(04), 367-383. doi: doi:10.1017/S0272263100008391
Univers
ity of
Mala
ya
133
Darus, S., & Khor, H.C. (2009). Common errors in written English essays of Form One
Chinese students: A case study. European Journal of Social Sciences, 10(2),
242-253.
Denzin, N.K. (2009). The research act: A theoretical introduction to sociological
methods: AldineTransaction.
Elliot, N. (1995). Narrative discourse and the basic writer. Journal of Basic Writing,
12(2), 19-30.
Ellis, Gail. (2014). ‘Young learners’: Clarifying our terms. ELT Journal, 68(1), 75-78.
doi: 10.1093/elt/cct062
Ellis, R. (1987). Interlanguage variability in narrative discourse: Style shifting in the use
of the past tense. Studies in Second Language Acquisition, 9(01), 1-19. doi:
doi:10.1017/S0272263100006483
Ellis, R. (2005). Planning and task performance in a second language. Amsterdam, The
Netherlands: John Benjamins.
Ellis, R. (2008). The study of second language acquisition: OUP Oxford.
Ellis, R. (2009). The differential effects of three types of task planning on the fluency,
complexity, and accuracy in L2 oral production. Applied Linguistics, 30(4), 474-
509.
Univers
ity of
Mala
ya
134
Ellis, R., & Barkhuizen, G.P. (2005). Analysing learner language: Oxford University
Press.
Ellis, R., & Yuan, F. (2004). The effects of planning on fluency, complexity, and
accuracy in second language narrative writing. Studies in Second Language
Acquisition, 26(01), 59-84.
Etikan, I., Musa, S. A., & Alkassim, S. (2016). Comparision of convenience sampling
and purposive sampling. American Journal of Theoretical and Applied Statistics,
5(1), 1-4.
Flower, L., & Hayes, J. R. (1981). A cognitive process theory of writing. College
Composition and Communication, 32(4), 365-387.
Foster, P., & Skehan, P. (1996). The influence of planning and task type on second
language performance. Studies in Second Language Acquisition, 18(03), 299-
323. doi: doi:10.1017/S0272263100015047
Galbraith, D. (2009). Cognitive models of writing. gfl-journal, 2(3), 7-21.
Gené-Gil, M., Juan-Garau, M., & Salazar-Noguera, J. (2015). Development of EFL
writing over three years in secondary education: CLIL and non-CLIL settings.
The Language Learning Journal, 43(3), 286-303. doi:
10.1080/09571736.2015.1053278
Univers
ity of
Mala
ya
135
Ghavamnia, M., Tavakoli, M., & Esteki, M. (2013). The effect of pre-task and on-line
planning conditions on complexity, accuracy, and fluency on EFL learner'
written production Porta Linguarum 20(3), 31-43.
Grabe, W., & Kaplan, R.B. (1996). Theory and practice of writing: An applied
linguistic perspective: Longman.
Graham, S., Harris, K. R., MacArthur, C. A., & Schwartz, S. (1991). Writing and
writing instruction for students with learning disabilities: Review of a research
program. Learning Disability Quarterly, 14(2), 89-114. doi: 10.2307/1510517
Graham, S., & Perin, D. (2007). A meta-analysis of writing instruction for adolescent
students. Journal of Educational Psychology, 99(3), 445-478. doi:
10.1037/0022-0663.99.3.445
Hacker, D.J., Dunlosky, J., & Graesser, A. C. (1998). Metacognition in educational
theory and practice: Taylor & Francis.
Heffner, C. L. (2015). Replication. Research Methods. Retrieved 24 November, 2015,
from http://allpsych.com/researchmethods/replication/
Henry, L. (2011). The development of working memory in children: SAGE Publications.
Hiew, W. (2012). English language teaching and learning issues in Malaysia: Learners'
perceptions via Facebook dialogue journal. Journal of Arts Science &
Commerce, 3(1).
Univers
ity of
Mala
ya
136
Holtzhausen, S. (2001). Triangulation as a powerful tool to strengthen the qualitative
research design: The resource-based learning career preparation programme
(RBLCPP) as a case study. Paper presented at the Higher Education Close Up
Conference 2, Lancaster University.
Housen, A., & Kuiken, F. (2009). Complexity, accuracy, and fluency in second
language acquisition. Applied Linguistics, 30(4), 461-473. doi:
10.1093/applin/amp048
Housen, A., Kuiken, F., & Vedder, I. (2012). Dimensions of L2 performance and
proficiency: Complexity, accuracy and fluency in SLA: John Benjamins
Publishing Company.
Hsu, H. (2012). Investigating the effects of planning on L2 text chat performance.
CALICO Journal, 29(4), 619-638.
Hudelson, S. (1989). Write on: Children writing in ESL. New Jersey: Center for
Applied Linguistics.
Hudson, R. (2009). Measuring maturity: SAGE Publications.
Hunt, K. W. (1970). Syntactic maturity in schoolchildren and adults. Monographs of the
Society for Research in Child Development, 35(1), iii-67. doi: 10.2307/1165818
Univers
ity of
Mala
ya
137
J., Mark D. (2012). Does planning really help? Effectiveness of planning in L2 writing.
Journal of Second Language Teaching and Research, 3(1), 107-118.
Jeon, H., Arlic, N., Brown, D., Lenz, K., Pusey, K., Rakita, K., . . . Wright, J. (2014).
Factors associated with second language writing. Northern Arizona University.
Johnson, M. D., Mercado, L., & Acevedo, A. (2012). The effect of planning sub-
processes on L2 writing fluency, grammatical complexity, and lexical
complexity. Journal of Second Language Writing, 21(3), 264-282. doi:
http://dx.doi.org/10.1016/j.jslw.2012.05.011
Kellogg, R. T. (1996). A model of working memory in writing. In C. M. Levy & S.
Ransdell (Eds.), The Science of Writing: Theories, Methods, Individual
Differences and Applications: Taylor & Francis.
Kellogg, R. T. (1999). Components of working memory in writing. In M. Torrance & G.
Jeffery (Eds.), The cognitive demands of writing: Processing capacity and
working memory effects in text production (pp. 43-61). Amsterdam: Amsterdam
University Press.
Kertas Soalan Peperiksaan Sebenar UPSR 2007-2013 UPSR Bahasa Inggeris. (2013).
Puchong, Selangor Darul Ehsan: Inspirasi Ilmu.
Kroll, B. (1990). Second language writing: Research insights for the classroom:
Cambridge University Press.
Univers
ity of
Mala
ya
138
Labov, W., & Waletzky, J. (1967). Narrative analysis: Oral versions of personal
experience. In A. E. Society & J. Helm (Eds.), Essays on the verbal and visual
arts: Distributed by the University of Washington Press.
Lanauze, M., & Snow, C. (1989). The relation between first-and second-language
writing skills: Evidence from Puerto Rican elementary school children in
bilingual programs. Linguistics and Education, 1(4), 323-339. doi:
http://dx.doi.org/10.1016/S0898-5898(89)80005-1
Latif, M. M. A (2009). It is more recursive than this: toward a new conceptualization of
the composing process. Paper presented at the Proceedings of the BAAL Annual
Conference 2009, Newcastle University.
Latif, M. M. M. A. (2012). What do we mean by writing fluency and how can it be
validly measured? Applied Linguistics. doi: 10.1093/applin/ams073
Li, L., Chen, J., & Sun, L. (2015). The effects of different lengths of pretask planning
time on L2 learners' oral test performance. TESOL Quarterly, 49(1), 38-66. doi:
10.1002/tesq.159
Lim, T. D. (2013). Analyzing Malaysian English classrooms: Reading, writing,
speaking and listening teaching strategies. University of Washington.
Lin, Y. (2013). The effects of task planning on L2 writing. (Doctor of Philosophy),
University of Auckland.
Univers
ity of
Mala
ya
139
Logie, R.H. (1995). Visuo-spatial working memory: Lawrence Erlbaum.
Lu, Y. (2010). Cognitive factors contributing to Chinese EFL learners' L2 writing
performance in timed essay writing. (Doctor of Philosophy), Georgie State
University.
Matsuda, P. K., Canagarajah, A. S., Harklau, L., Hyland, K., & Warschauer, M. (2003).
Changing currents in second language writing research: A colloquium. Journal
of Second Language Writing, 12(2), 151-179. doi:
http://dx.doi.org/10.1016/S1060-3743(03)00016-X
Mehnert, U. (1998). The effects of different lengths of time for planning on second
language performance. Studies in Second Language Acquisition, 20(01), 83-108.
Michel, M. C. , Kuiken, F., & Vedder, S. C. (2007). The influence of complexity in
monologic versus dialogic tasks in Dutch L2. IRAL, 45(2007), 241-259. doi:
10.1515/IRAL.2007.011
Ministry of Education Malaysia. (2001). Sukatan Pelajaran Kurikulum Bersepadu
Sekolah Rendah (Bahasa Inggeris).
Ministry of Education Malaysia. (2013a). Kurikulum Standard Sekolah Rendah:
Dokumen Standard Kurikulum Dan Pentaksiran Bahasa Inggeris Tahun Empat.
Univers
ity of
Mala
ya
140
Ministry of Education Malaysia. (2013b). SURAT PEKELILING IKHTISAS BIL. 2
TAHUN 2013: Pelaksanaan Kurikulum Standard Sekolah Rendah (KSSR)
Tahap II Secara Berperingkat-peringkat Mulai Tahun 2014. Putrajaya.
Ministry of Education Malaysia. (2015). UPSR Mulai Tahun 2016: Format Mata
Pelajaran Bahasa. Retrieved from http://www.moe.gov.my/cms/upload_files/art
iclefile/2015/articlefile_file_004254.pdf.
Mitchell, R., Myles, F., & Marsden, E. (2013). Second language learning theories:
Taylor & Francis.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T.
D. (2000). The unity and diversity of executive functions and their contributions
to complex “frontal lobe” tasks: A latent variable analysis. Cognitive Psychology,
41(1), 49-100. doi: http://dx.doi.org/10.1006/cogp.1999.0734
Moen, T. (2006). Reflections on the narrative research approach. International Journal
of Qualitative Methodology, 5(4).
Nam, H. H. (2011). A case study of an ESL child's writing development. Primary
English Education, 17(1), 383-394.
Norris, J. M., & Ortega, L. (2009). Towards an organic approach to investigating CAF
in instructed SLA: The case of complexity. Applied Linguistics, 30(4), 555-578.
doi: 10.1093/applin/amp044
Univers
ity of
Mala
ya
141
Norris, J., & Ortega, L. (2003). Defining and measuring SLA. In C. J. Doughty & M. H.
Long (Eds.), The handbook of second language acquisition: Wiley.
Oh, E. , Lee, C. M., & Moon, Y. (2015). The contributions of planning, L2 linguistic
knowledge and individual differences to L2 writing. The Journal of Asia TEFL,
12(2), 45-85.
Olive, T. (2004). Working memory in writing: Empirical evidence from the dual-task
technique. European Psychologist, 9(1), 32-42. doi: 10.1027/1016-9040.9.1.32
Olive, T., Kellogg, R. T., & Piolat, A. (2008). Verbal, visual, and spatial working
memory demands during text composition. Applied Psycholinguistics, 29(04),
669-687. doi: doi:10.1017/S0142716408080284
Olive, T., & Passerault, J. (2012). The visuospatial dimension of writing. Written
Communication, 29(3), 326-344. doi: 10.1177/0741088312451111
Ortega, L. (1999). Planning and focus on form in L2 oral performance. Studies in
Second Language Acquisition, 21(01), 109-148. doi: doi:null
Ortega, L. (2005). What do learners plan? Learner-driven attention to form during pre-
task planning. In R. Ellis (Ed.), Planning and task performance in a second
language. Amsterdam, The Netherlands: John Benjamins.
P., Julie. (2005). SPSS survival manual. Sydney: Allen & Unwin.
Univers
ity of
Mala
ya
142
Paivio, A. (1986). Mental representations: A dual coding approach: Oxford University
Press.
Pallotti, G. (2009). CAF: Defining, refining and differentiating constructs. Applied
Linguistics, 30(4), 590-601. doi: 10.1093/applin/amp045
Patil, R. (2014). Task based language learning. Research Front, 1(Special Issue), 365-
368.
Piri, F., Barati, H., & Ketabi, S. (2012). The effects of pre-task, on-line and both pre-
task and on-line planning on fluency, complexity and accuracy - The case of
Iranian EFL learners' written production. English Language Teaching, 5(6), 158-
167. doi: 10.5539/elt.v5n6p158
Rajathurai, P. (2016). Kertas Soalan Peperiksaan UPSR Bahasa Inggeris. Ampang,
Selangor Darul Ehsan: Laman Klasik Resources.
Rahimpour, M., & Nariman-Jahan, R. (2011). The effects of planning on writing
narrative task performance with low and high EFL Proficiency. English
Language Teaching, 4(1), 120-127.
Richards, J. C., & Rodgers, T. S. (2001). Approaches and methods in language teaching:
Cambridge University Press.
Richards, J. C., & Schmidt, R. W. (2002). Longman dictionary of language teaching
and applied linguistics: Longman Publishing Group.
Univers
ity of
Mala
ya
143
Rozovskaya, A., Roth, D., & Srikumar, V. (2014). Correcting grammatical verb errors.
Paper presented at the EACL.
Rungruangthum, M., Todd, R. W., & Aroonmanakun, W. (2011, 21-22 April). Re-
conceptualization of piloting research in applied linguistics. Paper presented at
the Doing Research in Applied Linguistics, Bangkok: King Mongkut's
University of Technology Thonburi.
Salimi, A., Parviz, A., & Hosseini, P. (2012). The effect of strategic planning time and
task complexity on L2 written accuracy. Theory and Practice in Language
Studies, 2(11), 2398-2406. doi: 10.4304/tpls.2.11.2398-2406
Sangarun, J. (2001). The effects of pre-task planning on foreign language performance.
(Doctor of Philosophy), University of Toronto.
Sangarun, J. (2005). The effects of focusing on meaning and form in strategic planning.
In R. Ellis (Ed.), Planning and task performance in a second language.
Amsterdam: John Benjamins.
Schacter, D.L., Wagner, A.D., & Bucker, R.L. (2000). Memory Systems of 1999. In E.
T. R. S. R. R. Institute & F. I. M. C. P. P. U. Toronto (Eds.), The Oxford
Handbook of Memory: Oxford University Press, USA.
Scott, K. (2014). 7-11 planning sheets. Retrieved 30/07/2014, 2014, from
http://www.teachingideas.co.uk/english/files/planningsheets2.pdf
Univers
ity of
Mala
ya
144
Seifollahi, M., & Tamjid, N. H. (2012). The effect of mixed planning on the fluency and
complexity of EFL learners’ writing performance. Procedia - Social and
Behavioral Sciences, 46(0), 764-770. doi: http://dx.doi.org/10.1016/j.sbspro.2012
.05.195
Seyyedi, K., Mohamed, I. S. A. M., Orang, M., & Nejad, S. M. (2013). The effect of
pre-task planning time on L2 learners' narrative writing performance. English
Language Teaching, 6(12), 1-9. doi: doi:10.5539/elt.v6n12p1
Shi, L. (1998). Effects of prewriting discussions on adult ESL students' compositions.
Journal of Second Language Writing, 7(3), 319-345. doi:
http://dx.doi.org/10.1016/S1060-3743(98)90020-0
Shin, Y. (2008). The effects of planning on L2 writing: a study of Korean learners of
English as a foreign language. (PhD (Doctor of Philosophy)), University of
Iowa. Retrieved from http://ir.uiowa.edu/etd/44.
Silva, T. (1993). Toward an understanding of the distinct nature of L2 writing: The ESL
research and its implications. TESOL Quarterly, 27(4), 657-677. doi:
10.2307/3587400
Sim, B. Y. (2010). The effects of planning on the narrative writings of Malaysian
tertiary level students. (MESL), University of Malaya.
Skehan, P. (1998). A cognitive approach to language learning: OUP Oxford.
Univers
ity of
Mala
ya
145
Skehan, P. (2009). Modelling second language performance: Integrating complexity,
accuracy, fluency, and lexis. Applied Linguistics, 30(4), 510-532. doi:
10.1093/applin/amp047
Skehan, P., & Foster, P. (1997). Task type and task processing conditions as influences
on foreign language performance. Language Teaching Research, 1(3), 185-211.
doi: 10.1177/136216889700100302
Skehan, P., & Foster, P. (1999). The influence of task structure and processing
conditions on narrative retellings. Language Learning, 49(1), 93-120. doi:
10.1111/1467-9922.00071
So, L. (2015). L2 writing instruction in blended learning for the development of fluency,
complexity, and accuracy in higher education. Multimedia-Assisted Language
Learning, 18(4), 121-147.
Soleimani, H., & Kargozari, H. R. (2014). Uncovering new trends of investigation in
written task planning. International Journal of Asian Social Science, 4(9), 991-
1002.
Stainthorp, R. (2004). What is involved in writing? Understanding the processes that
enable children to become fluent writers: A psychological perspective. Draft
Paper. Institute of Education, University of London.
Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach's alpha. International
Journal of Medical Education, 2, 53-55. doi: 10.5116/ijme.4dfb.8dfd
Univers
ity of
Mala
ya
146
Tavakoli, P., & Skehan, P. (2005). Strategic planning, task structure, and performance
testing. In R. Ellis (Ed.), Planning and task performance in a second language.
Amsterdam, The Netherlands: John Benjamins.
Ting, H. M. H. (2013). Language, identity and mobility: Perspective of Malaysian
Chinese youth. Malaysian Journal of Chinese Studies, 2(1), 83-102.
Van Dijk, T. A. (1976). Narrative macro-structures: Logical and cognitive foundations.
PTL: A Journal for Descriptive Poetrics and Theory of Literature 1, 547-568.
Van Dijk, T.A. (1977). Text and context: Explorations in the semantics and pragmatics
of discourse: Addison-Wesley Longman Limited.
Vercelloti, M. L. (2012). Complexity, accuracy, and fluency as properties of language
performance: The development of the multiple subsystems over time and in
relation to each other. (Doctor of Philosophy), University of Pittsburgh,
Pittsburgh.
Waes, V. L., & Leijten, M. (2015). Fluency in writing: A multidimensional perspective
on writing fluency applied to L1 and L2. Computers and Composition, 38, Part
A, 79-95. doi: http://dx.doi.org/10.1016/j.compcom.2015.09.012
Ward, T. (2007). Re-gendering data: Quantifying qualitative. Paper presented at the
Association for Institional Research 2007 Annual Forum.
Univers
ity of
Mala
ya
147
Wendel, J. N. (1997). Planning and second-language narrative production. (Doctoral
Degree), Temple University, Philadelphia, Pennysylvania.
Wigglesworth, G. (1997). An investigation of planning time and proficiency level on
oral test discourse. Language Testing, 14(1), 85-106. doi:
10.1177/026553229701400105
Williams, J. (1992). Planning, discourse marking, and the comprehensibility of
international teaching assistants. TESOL Quarterly, 26(4), 693-711. doi:
10.2307/3586869
Yi, B., & Luo, S. (2013). Working memory and lexical knowledge in L2 argumentative
writing. Asian Journal of English Language Teaching, 23, 83-103. doi:
10.1353/aje.2013.0004
Yi, B., & Ni, C. (2015). Planning and working memory effects on L2 performance in
Chinese EFL learners' argumentative writing. Indonesian Journal of Applied
Linguistics, 5(1), 43-51. doi: http://dx.doi.org/10.17509%2Fijal.v5i1.830
Yuan, F., & Ellis, R. (2003). The effects of pre‐task planning and on‐line planning on
fluency, complexity and accuracy in L2 monologic oral production. Applied
Linguistics, 24(1), 1-27.
Zimmermann, R. (2000). L2 writing: subprocesses, a model of formulating and
empirical findings. Learning and Instruction, 10(1), 73-99. doi:
http://dx.doi.org/10.1016/S0959-4752(99)00019-5
Univers
ity of
Mala
ya